# HSEF v5.3 review: Neuroinflammation Hypothesis

> Sources: 69 primary reports; claims: 74; verified: 74; unverified: 0. Verification is bounded attribution. Access: 6 partial-body and 63 complete abstract/status examinations. Literature cutoff: 2026-10-06; source examination UTC date: 2026-10-06. No complete-paper audit or blinded calibration claimed.

> **Adversarial review:** Generated by Sol 6.1; adversarially reviewed by Claude Opus 5.5 (Anthropic) on 2026-10-06. 3 corrections applied (see ADVERSARIAL REVIEW AND APPROVAL). Status: **Approved as revised.**

## AT A GLANCE

**Glial states can causally amplify degeneration, but harmful activation is not a general rule across disease, stage and human intervention.**

| Level | Confidence | Causal role | Momentum | Stability | Divergence |
|---|---|---|---|---|---|
| Γ — Mechanistically plausible, partial system support | Moderate | reinforcing / stabilizing factor | Mechanistic only | Stable | Contested |

**Deciding experiment:** Induce loss of a prespecified harmful glial program after comparable neuronal pathology begins, preserve protective clearance and peripheral immune function, then restore the same mediator with cell-specific rescue in independent AD, PD and ALS settings.

**Belief versus evidence:** The registered weakest-link conjunction returns Γ because disease-class human generalization remains unresolved. Defined glial states have real causal effects on neuronal loss and progression, including reactive amplifiers. This evidence is not dismissed for being downstream. Opposite effects of protective immune programs and negative or harmful human interventions prevent treating chronic activation as one broadly pathogenic exposure. The limiting gate is generalization and state-specific human mediation, not an added obligation to initiate disease.

**Best-supported rival:** Reactive but causal amplifier. Use post-onset cell-state-selective loss/rescue with the initiating neuronal lesion held comparable. Measure neuron survival and function as well as clearance and peripheral immunity. A reactive amplifier predicts progression change without obligatory onset prevention; a purely passive response predicts no such change.

## COMPETING MECHANISTIC MODELS

**Rank 1: Reactive but causal amplifier (Ε).** Defined glial states respond to proteinopathy and causally worsen subsequent degeneration. SOD1 glial removal, NFκB suppression and cytokine deletion slow later progression; APOE4 and immune-hub perturbations protect tau models. Shared mutant backgrounds, systemic effects and uncertain human mediation prevent broader unconditional claims.

**Rank 2: Protective clearance and immune failure (Ε).** Some activated states preserve tissue through clearance, containment and immune coordination. SYK, TREM2, IL3, TRIF and MyD88 contrasts show deterioration when protective responses are impaired. The same receptor can have opposite effects with amyloid versus tau, APOE genotype or timing.

**Rank 3: Neuron-autonomous proteinopathy (Ε).** Intrinsic neuronal protein dysfunction initiates injury that glia then modify. Motor-neuron SOD1 controls onset/early progression whereas glial removal primarily changes later progression. It does not explain all progression without glial contributions and is not a universal origin claim.

**Rank 4: Peripheral immune contribution (Ε).** Adaptive and peripheral innate signals interact with glia and alter defined model trajectories. T-cell depletion, IFNγ blockade, marrow MyD88 and immune-suppression contrasts affect atrophy or survival. Systemic interventions and brief human immune-modulation signals do not isolate a universal mediator.

**Rank 5: Primary chronic glial driver (Γ).** Chronic harmful glial activation broadly drives degeneration. Causal perturbations across model systems and human immune-risk genetics support disease-relevant participation. Broad activation is heterogeneous; causal human generalization and selective state mediation are unresolved.

The tournament does not alter the nominal level. Compatible pathways and flexible alternatives are not automatically proved by explaining observations.

## FULL REVIEW

### Registered claim and scope

Chronic microglial and astrocytic activation is a causal driver of neurodegeneration rather than a reactive response to it.

Retain the original causal-driver-versus-reactive wording without silently adding an obligatory inflammation-first initiation claim. Presence, causal contribution, direction and cross-disease scope are scored separately. A downstream amplifier may be causal despite being reactive; this semantic distinction is explicit.

Selection: public-interest rank 23 from the editorial ranking compiled 2026-09-23; categorical prior verdicts were visible. Private fresh review; earlier documents are preserved.

### Evidence and causal interpretation

Early tissue, human preclinical fluid and single-cell observations locate glial responses along disease trajectories, but postmortem reconstruction and structural equation modeling do not establish origin. Human genetic immune signals are disease relevant while allowing both protective and harmful functions (PMID39402379, PMID40747577, PMID23150934, PMID23150908, PMID28714976).

SOD1 removal distinguishes neuronal onset control from glial progression modification. Microglial-lineage partial NFκB deletion extends median survival without changing onset or mutant SOD1 abundance. Homozygous deletion causes serious immune dysfunction, and the CSF1R lineage is not restricted to CNS microglia. The causal late effect remains real within this context (PMID16741123, PMID18246065, PMID24607225).

Global IL1α/TNF/C1q triple deletion delays motor-neuron loss and denervation, with similar eventual loss at a later endpoint. Inducible astrocytic APOE4 deletion after initial tau pathology protects atrophy, NfL and nest building; APOE3 and Cre-negative tamoxifen contrasts constrain interpretation. Neither experiment isolates a universally harmful activation state (PMID32719333, PMID33831349).

Microglia–T-cell hub perturbations reduce tau-model atrophy and function defects. Original Results verify T-cell depletion, IFNγ inhibition and late PLX3397 timing, but tau burden changes and PLX3397 has CSF1R/c-kit/FLT3 targets. Peripheral immunity and glial signals interact rather than defining one exclusive initiator (PMID36890231).

Protective immune biology is substantial: SYK deletion lowers plaque-associated activation while worsening Aβ, neuritic damage and memory; CLEC7A activation restores uptake/plaque features without a demonstrated long-term behavioral rescue. TREM2 effects reverse across tau/amyloid/APOE contexts. IL3, TRIF and marrow MyD88 similarly preclude a simple suppression rule (PMID36306735, PMID34262178, PMID29073081, PMID32544086, PMID36368315, PMID29568058, PMID18086918).

PD inflammasome and STING experiments support causal model participation, whereas complement and necroptosis nulls constrain particular mediators. Activated astrocytes can also assist aggregate removal in culture. Cell-state labels, lysosomal markers and TH staining require their own measurement limits (PMID30381407, PMID35394877, PMID17900537, PMID24170856, PMID30815534, PMID31745214, PMID33046546, PMID39059388).

Human results are not uniformly encouraging: AL002 has CSF target engagement but no clinical benefit and frequent ARIA; NLY01 and AZA-PD miss primary endpoints; two semaglutide trials are clinically null. Minocycline accelerates ALS functional decline without a statistically established mortality excess. Broad or incompletely selective drugs constrain their tested strategies, not every inflammatory mechanism (PMID41787076, PMID38101901, PMID41389828, PMID41865758, PMID17980667).

ADAPT prevention and symptomatic NSAID trials are null. Sargramostim short safety trials have exploratory signals: the AD study has baseline MMSE imbalance, mixed cognitive outcomes and null function/PET, while the PD study is small and brief. NT0796 offers short open-label CSF engagement, not proven disease modification (PMID17460158, PMID23562431, PMID12783912, PMID33778150, PMID28649610, PMID40792655).

Consensus sharpened the initiation/amplification distinction and retrieved indexed experimental passages. Its targeted final response supplied Boillee at abstract level and Yamanaka, Guttenplan and Chen as body-level extractions, contradicting an interim all-four full-text claim. Named trial retrieval failed; direct primary reports filled that gap. Only passages checked against original sources receive full-text validation credit.

## HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

<!-- consensus-consistency-2026-10-09 -->
### Consensus provenance reconciliation — 9 October 2026

Case-specific Consensus discovery records were located and linked to this final edition. This supplements the review's provenance; it leaves the original assessed claim, grades and source counts intact. Provider summaries and full-text badges do not establish examination of original primary passages by this reviewer. Source access and incorporation retain the limits stated in the original methods and any dated research update.

Recorded evidence artifacts:

- `deliverables/hsef_v53_popularity_ranks_21_25_2026-10-06/neuroinflammation-ranked-refresh/consensus/broad-browser-snapshot.txt`
- `deliverables/hsef_v53_popularity_ranks_21_25_2026-10-06/neuroinflammation-ranked-refresh/consensus/query-ledger.json`
- `deliverables/hsef_v53_popularity_ranks_21_25_2026-10-06/neuroinflammation-ranked-refresh/consensus/reconciliation.md`
- `deliverables/hsef_v53_popularity_ranks_21_25_2026-10-06/neuroinflammation-ranked-refresh/consensus/targeted-browser-snapshot.txt`

The located artifacts do not state a reliable source-comparison count. No count is inferred.

A dated adversarial approval record exists for the separately identified neuroinflammation-ranked-refresh.review.md (2026-10-06; see edition-approval-audit.json). Its scope is that recorded edition. This supplement has no separate approval and no coverage is inferred from topic identity. No score or confidence change is made by this reconciliation.
<!-- end-consensus-consistency -->


### Subclaim matrix and aggregation

| Subclaim | Scope | Level | Decision |
|---|---|---|---|
| SC1 Glial activation is present in neurodegenerative disease and can precede overt neuronal loss. | AD, PD and ALS with prespecified stage and cell type | Γ | Presymptomatic animal changes and human preclinical/longitudinal observations exceed terminal colocalization alone. Human state ordering and direct neuronal-loss mediation remain unresolved; pseudotime and structural equation modeling do not identify an initiating event. |
| SC2 Specific chronic glial states contribute causally to neuronal loss and functional decline. | defined cell states, genotypes, stages and disease models | Ε | Endogenous cell-lineage perturbations of mutant SOD1, NFκB and astrocytic APOE, as well as complementary immune manipulations, materially alter survival, atrophy or function in specified models. Defined mediator causality crosses Δ and merits Ε with explicit genotype, state and timing restrictions. Protective SYK, TREM2, IL3 and innate signaling states rule out indiscriminate activation suppression. |
| SC3 Glial activation drives degeneration beyond a purely reactive or clearance response. | disease-relevant temporal and bidirectional perturbation settings | Ε | Glial contributions exceed a passive bystander or clearance-only account. SOD1 removal in glia and partial NFκB suppression slow later progression without changing onset; these are reactive causal amplification, which qualifies under the registered wording. Astrocyte and microglial effects are not interchangeable and do not establish obligatory upstream initiation. This defined causal contribution is Ε. |
| SC4 The causal-driver claim generalizes across AD, PD and ALS. | disease-class human and model populations defined before outcomes | Γ | AD, PD and ALS models show context-dependent causal immune effects, but human pathway engagement has not reliably produced sustained clinical benefit. AL002, NLY01, AZA-PD, ADAPT, NSAID treatment and minocycline test different, incompletely selective mechanisms. Protective and harmful states require prospective boundaries; the general disease-class claim remains Γ. |

Preregistered aggregation: weakest-link. The registered weakest-link conjunction returns Γ because disease-class human generalization remains unresolved. Defined glial states have real causal effects on neuronal loss and progression, including reactive amplifiers. This evidence is not dismissed for being downstream. Opposite effects of protective immune programs and negative or harmful human interventions prevent treating chronic activation as one broadly pathogenic exposure. The limiting gate is generalization and state-specific human mediation, not an added obligation to initiate disease. Sensitivities: weakest-link Γ; primary-subclaim Ε; scope-matched Ε.

### Decisive-perturbation inventory and causal tally

[2] Astrocytic IL-3 programs microglial motility and aggregate clearance and restricts pathology and cognitive decline in tested AD models. Protective activation contradicts treating activation as uniformly toxic; constitutive cytokine and model backgrounds limit natural human mediation.

[4] Parkin loss increased LPS-triggered microglial NLRP3 assembly, motor impairment and dopaminergic loss; Parkin overexpression and HSP90 inhibition interrogated NLRP3 degradation. Global Parkin deletion, imposed systemic LPS and broad HSP90 effects do not isolate a microglial-only natural PD initiator.

[5] Microglia-specific IL-10 delivery protected dopaminergic neurons in SNCA-overexpressing mice while inducing an activated prophagocytic microglial state and lymphocyte regulation. Activation remains present in a protective state; local cytokine effects extend beyond microglia, and imposed SNCA expression restricts scope.

[6] NLY01 reduced dopaminergic loss and behavioral deficits after synuclein fibrils and prolonged survival in A53T mice, associated with reduced neurotoxic astrocyte conversion. GLP1R agonism and forced/transgenic proteinopathy do not establish selective glial mediation or human efficacy; the clinical trial is assessed separately.

[7] MCC950 inhibited NLRP3 activity and improved motor, dopamine-neuron and aggregate endpoints in several rodent PD models; degeneration itself could activate the pathway. This supports an injury-triggered causal amplifier, not obligatory inflammation-first origin; drug/pathway and induced-model scope must be retained.

[8] C1q deletion reduced microglial and astrocytic synapse engulfment and restored density in TauP301S; astrocytic engulfment compensated when Trem2-dependent microglial phagocytosis failed. Synapse density is not total neuronal survival, and global complement deletion affects multiple functions and engineered backgrounds.

[9] A single high-dose systemic LPS or TNF exposure caused sustained brain inflammation and delayed progressive TH-positive nigral-neuron loss, with TNF-receptor comparisons. The imposed systemic inflammatory insult and receptor-wide deletion do not establish native chronic glial primacy; TH loss is not an audited all-neuron count.

[10] Human iPSC astrocytes reduced neuronal synuclein accumulation and transfer through lysosomal clearance; ATP13A2-deficient astrocytes lost part of this protection. Protective glial function is demonstrated in culture, not intact organismal progression or a general absence of harmful glial states.

[11] Cx3cr1 deficiency worsened AAV-A53T synuclein inflammation and dopaminergic degeneration, while wild-type synuclein injury was similar across receptor genotypes. Receptor loss can remove restraint rather than simply reduce activation; developmental and mutant-expression contexts limit broad extrapolation.

[12] C1q, C3 and CR3 inhibition reduced early amyloid-associated synapse loss; C1q was required for oligomer-induced synaptic and LTP toxicity in the tested models. Early means before overt plaques, not before all amyloid or neuronal stress; complementary pathway controls support conditional mediation without proving human initiation.

[13] TREM2 deficiency impaired microglial metabolism, and cyclocreatine restored plaque clustering while reducing adjacent neuronal dystrophy. A protective energetic/clearance program can be activated; metabolic rescue is broader than a single inflammatory mediator and occurs in amyloid-model scope.

[14] Microglial SYK deficiency impaired plaque encasement and worsened pathology and behavior; CLEC7A activation rescued microglial activation in TREM2-R47H mice. Different activated programmes have different effects; antibody rescue of activation is not independently established complete human clinical rescue. Original Results: inducible SYK deletion lowers plaque-associated microgliosis but increases Aβ, neuritic dystrophy and memory impairment. Monocytes recover after washout; border-associated macrophages retain partial deletion. Late deletion also impairs clustering. CLEC7A rescue restores uptake and plaque morphology, not demonstrated long-term behavioral rescue.

[15] TREM2-C1q binding restrained complement activation; haploinsufficiency worsened synapse engulfment and a binding peptide rescued synaptic impairment in tested models. Human tissue complex associations do not establish causal direction; complement-restraint protection is distinct from globally harmful activation.

[16] Nptx2 loss increased C1q-dependent microglial synapse removal, while neuronal Nptx2 overexpression restrained complement and synapse loss in aged tau mice. The initiating alteration is neuronal complement restraint and one rescue uses AAV overexpression; synapse preservation is not whole-disease neuronal survival.

[17] TREM2 loss impaired damage-lipid sensing, microglial plaque clustering and survival and increased amyloid accumulation in 5xFAD. Protective immune responsiveness is demonstrated in an amyloid context; amyloid load alone does not establish cognitive/neuron-loss mediation or chronic-state universality.

[18] Myeloid or microglial LANDO loss increased cytokines and neurotoxic amyloid, with accelerated degeneration, altered signaling and memory deficits. Loss of aggregate clearance may precede harmful inflammation, so pure inflammatory primacy is not isolated; a linked correction remains qualified.

[19] Human TREM2-R47H in PS19 reduced microglial reactivity, synapse phagocytosis, atrophy and synapse loss versus the common variant. An AD-risk allele is protective in this tau-only model, mapping upstream substrate/stage context rather than proving uniform effects in natural AD.

[20] Open-label NT-0796 in ten PD participants for 28 days showed CSF exposure and reduced inflammatory markers, with four healthy volunteers also studied. Short safety/pharmacokinetic biomarker engagement without randomized clinical trajectory is not disease modification or evidence of a glial initiating event.

[22] C7 antibody and C6 knockout reduced plaque-adjacent spine/synapse loss in two AD models, implicating terminal complement activity. Synapse endpoints are not audited neuron-survival/function trajectories, and constitutive knockout or systemic antibody may affect other compartments.

[23] Microglia-selective synuclein overexpression caused dopamine-neuron degeneration without neuronal synuclein aggregation, and oxidative/nitrosative inhibition attenuated damage. A strong imposed glial sufficiency setting is not physiological endogenous primacy; peripheral recruitment and broad oxidative inhibition remain alternative mediators. Linked correction qualified.

[24] STING activation preceded dopamine-neuron loss after synuclein fibrils; STING-deficient mice resisted motor, aggregate and neuronal-loss phenotypes. Global STING deficiency and imposed fibrils constrain cell specificity and natural initiation; human staining is correlational.

[25] Inducible astrocytic APOE4 removal after tau pathology onset reduced neurodegeneration, synapse loss and microglial synaptic phagocytosis. Post-onset protection supports causal amplification; APOE affects lipid and tau pathways in addition to inflammation, and the P301S/APOE context remains explicit. Original Results: astrocytic APOE deletion at 5.5 months after initial tau pathology protects APOE4 but not APOE3 mice from atrophy. Tamoxifen-treated Cre-negative controls exclude tamoxifen alone; NfL, nest building and synaptic measures improve. APOE influences lipid handling and tau pathology as well as inflammatory states, so the mediation is not inflammation-selective.

[27] Intracerebral clodronate depleted microglia but also damaged vascular integrity, reduced neural markers and increased inflammatory/glial signals; neuronal ATP fell in culture. This methodological perturbation exposes depletion off-target confounding, rather than showing that all microglia protect in AD, PD or ALS.

[28] Microglial IL1a, TNF and C1q induced toxic reactive astrocytes; blocking formation protected axotomized neurons, and human disease tissue contained the state. Injury precedes the causal astrocyte response in the in-vivo test; human postmortem abundance and a marker class do not establish obligatory inflammation-first disease origin.

[29] Single-cell mapping and Trem2 loss separated two disease-associated microglial activation phases, with intracellular amyloid in mapped cells. Potential protective function is not a selective hard-outcome rescue in this abstract; the DAM name is not a causal toxicity classification.

[30] Microglial nanotube contacts removed neuronal tau/synuclein aggregates, supplied healthy mitochondria and restored activity; mitochondrial poisoning removed protection. Broad mitochondrial perturbation and culture-based transfer do not establish organismal disease trajectory; aggregate transfer can be protective rather than uniformly toxic.

[31] NLRP3 loss reduced tau phosphorylation/aggregation, and amyloid-homogenate-induced tau pathology depended on NLRP3; tau itself activated the inflammasome. The test supports a bidirectional proteinopathy-immune pathway, with imposed initiation and pathology endpoints rather than a complete human survival dependency.

[33] Microglial PD1 deletion increased inflammation and impaired amyloid uptake, with more plaque deposition in APP/PS1. Loss of immune restraint and clearance is harmful in this context; plaque load is not independently demonstrated neuron loss or general inflammatory primacy.

[34] ASC specks promoted amyloid cross-seeding after hippocampal delivery, and ASC deficiency or antibody coapplication blocked induced pathology. Forced amyloid/speck entry and aggregate endpoints establish inflammatory seeding capability, not native disease initiation or clinical benefit.

[35] AL002c stimulated microglial proliferation and reduced filamentous plaques and neurite dystrophy in a human-TREM2 AD model; first-human fluid engagement was also reported. Activation can be protective in this model, and phase1 engagement is not human clinical efficacy; later reports overlap the clinical programme.

[36] TREM2 knockout worsened tau pathology and degeneration in P301S/APOE4 despite reduced TREM2-dependent microgliosis. This reverses the earlier mouse-Apoe tau-only protection contrast, showing genotype and residual microglial state matter more than bulk activation reduction.

[37] INVOKE1 in 64 healthy volunteers showed AL002 CSF penetration, soluble-TREM2 reduction and signaling/recruitment biomarker increases with acceptable single-dose tolerability. Healthy phase1 safety and engagement cannot establish AD trajectory benefit; the 2020 programme report is not independent efficacy evidence.

[38] The AD-protective PLCG2-P522R knock-in increased enzyme activity, macrophage survival/acute responses and measures of microglial function. Protective genetic association is from a separate study; cell function and PET signals do not prove chronic harmful activation or disease rescue in this KI experiment.

[44] A 20-person randomized PD phase1 study reported modest short-term motor improvement with sargramostim, peripheral Treg changes and substantial adverse-event burden. Safety-focused small enrollment, short exposure and peripheral modulation do not establish sustained neuroprotection or prove a CNS glial mediator.

[45] Heterozygous TBK1 loss advanced SOD1 onset/denervation but reduced late microglial inflammation, slowed progression and extended survival. Global pleiotropic autophagy/immune effects show stage-dependent direction, rather than a universal beneficial inflammatory suppression or native sporadic ALS cause.

[46] xCT deletion reduced microglial glutamate and inflammatory outputs and slowed late SOD1 degeneration with more surviving motor neurons, despite earlier symptom onset. Global antiporter deletion and redox effects complicate microglial specificity; onset and progression move in opposite directions.

[48] Conditional mutant-SOD1 reduction in astrocytes left onset unchanged but delayed microglial activation and sharply slowed later progression. This is direct cell-resolved causal amplification in familial SOD1 models, not proof of inflammation-first initiation or every reactive astrocyte being toxic.

[49] Cell-resolved mutant-SOD1 deletion separated motor-neuron-determined onset/early progression from microglial effects on late progression. Different disease phases and engineered familial mutation define scope; reducing mutant protein may restore several glial functions beyond inflammation.

[50] Astrocytic TGFβ1 overproduction worsened SOD1 progression while deactivating protective microglia/T cells; post-onset signaling inhibition extended survival. Reducing one cytokine can improve protective immune activity rather than uniformly suppress activation; source overproduction and drug pleiotropy remain limits.

[51] TRIF deficiency shortened SOD1 survival with accumulated abnormal astrocytes; manipulating selected lymphocyte populations did not substantially alter survival. Innate signaling can protect the motor-neuron environment, and cell correlations do not by themselves prove which abnormal astrocyte state mediates the whole effect.

[52] Microglial NFκB suppression extended ALS-mouse survival, whereas astrocytic suppression did not rescue motor-neuron death; microglial activation caused death in culture and in vivo. The endogenous pathway perturbation is cell-resolved, but forced constitutive activation and SOD1 genotype restrict broad human or inflammation-first inference. Original Results: CSF1R-Cre acts in the mononuclear phagocyte lineage, although spinal reporter staining is restricted to microglia. Homozygous IKKβ deletion caused serious immune dysfunction, so survival was assessed in heterozygotes. Median survival rose from 133 to 153 days, onset was unchanged, and mutant SOD1 abundance was unchanged. Forced NFκB activation also kills neurons, but is not physiological initiation.

[53] NLRP3 or caspase1 loss preserved spatial memory and increased amyloid clearance in APP/PS1, with reduced inflammatory activity and amyloid burden. Constitutive global immune loss and familial amyloid background are limited scopes; downstream response to amyloid can still mediate causal decline.

[54] One-month CSF1R inhibition in ten-month 5xFAD removed most microglia and preserved neuronal/spine endpoints and contextual memory without reducing amyloid. This is a post-pathology causal contribution; systemic receptor pharmacology and total-cell removal are broader than a single harmful state, and protective contexts elsewhere remain.

[61] A 40-person randomized AD study found short-term MMSE improvement after three weeks of sargramostim, with peripheral immune and plasma biomarker changes. Safety-primary small study and brief cognitive effects do not establish sustained disease modification or a CNS glial mechanism; broader outcome and baseline details require original sections. Original Results and Methods: 40-person, three-week safety-primary trial. Baseline MMSE imbalance complicates the positive MMSE signal; baseline-adjusted end-of-treatment significance is lost, although the first follow-up remains positive. ADAS-Cog13 worsens at that follow-up, ADL/CDR-SB are null and the small PET subset is null. This is a mixed exploratory efficacy pattern, not established sustained benefit.

[63] Normal astrocyte-precursor transplantation in SOD1 rats reduced microgliosis, preserved motor neurons and respiratory/motor function and extended survival, partly through GLT1. Transplantation restores trophic/glutamate functions as well as inflammation; focal familial-model benefit does not isolate one harmful reactive state.

[64] AraC removal of proliferating CNS cells accelerated SOD1 progression while reducing microglia, progenitors and T cells. Multiple cell classes and anti-mitotic effects prevent attributing harm specifically to microglial loss, but contradict indiscriminate suppression as a universal strategy.

[66] MyD88-deficient marrow transplantation advanced onset and shortened survival in SOD1-G37R mice relative to competent donor cells. Marrow conditioning and peripheral/engrafted myeloid contributions differ from native resident microglial activation; immune signaling was protective in this tested context.

[67] Combined IL1a, TNF and C1q knockout extended SOD1-G93A survival and interrogated the proposed microglia-to-reactive-astrocyte pathway. Global lifelong cytokine deletion is not astrocyte-specific loss; rescue, stage timing and alternative immune effects require body checks before attributing all benefit to one state. Original Results: lifelong global IL1α/TNF/C1q triple deletion delays motor-neuron loss and denervation rather than preventing eventual disease; mutant animals reach a later endpoint with similar loss. Disease onset remains similar. This is causal progression modification with a constitutive, systemic intervention.

[68] Microglial or T-cell depletion blocked tau-model degeneration, while IFNγ or PDCD1 inhibition ameliorated atrophy; human AD tissue contained expanded cytotoxic T cells. Adaptive and innate immune effects form a causal model hub downstream of tau, rather than proving all human activation is an initiating driver; linked correction remains attached. Original Results: peripheral T-cell depletion from 6 to 9.5 months reduces atrophy, NfL and functional deficits versus IgG controls; IFNγ blockade begins at 7.5 months and PLX3397 at 8.5 months. PLX3397 targets CSF1R, c-kit and FLT3. Tau burden also changes, and neither intervention isolates a glial-only mediator. Original results verify better Y-maze/auditory-cue performance and baseline locomotor controls; contextual freezing was only a trend.

[69] Trem2 deletion reduced microgliosis and preserved brain volume in PS19 despite similar phosphorylated/insoluble tau. This separates tau burden from immune-associated atrophy in a pure-tau background; amyloid and APOE4 experiments reverse or alter the direction, restricting generalization.

Tally: 50 discriminating contrasts. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Shared cohorts, experimental platforms, transgenic lineages and repeated outcomes do not constitute fully independent causal lines.

### Cap audit, precedence and Β/Γ/Δ boundary

The primary neurodegeneration and secondary immunology rubrics agree that endogenous mediator perturbation supports causal contribution in the tested context. Mutant SOD1 or tau overexpression alone is not strong causal evidence; mediator deletion, timing and survival/function contrasts provide an additional causal test. That distinction permits Ε for bounded SC2/SC3 while leaving broad SC4 at Γ. Protective immune-state interventions count as contradictions to a pan-activation claim, not as grounds to deny all inflammatory causality. Independent convergence is assessed as a set, without imposing an impossible universal RCT.

Gamma hard-rule audit: Pure histology and biomarker association alone cannot establish direction. Endogenous mediator deletion in a mutant disease background supports the tested context, not automatic sporadic-human generalization. Reactive causal amplification is distinguished from both passive response and an obligatory initiator. The assembled set is assessed for complementary failure modes; trial count or source count is not convergence.

SC1 field-valid definition: “
- Endogenous genetic deletion, suppression, or degron depletion of the proposed mediator in vivo
- Selective pathway inhibition that changes onset/progression while preserving broader viability
- Vagotomy-type natural experiment, only if confounding is rigorously controlled AND effect is disease-initiation relevant rather than merely associated” Positive full-scope result: no. Presymptomatic animal changes and human preclinical/longitudinal observations exceed terminal colocalization alone. Human state ordering and direct neuronal-loss mediation remain unresolved; pseudotime and structural equation modeling do not identify an initiating event.

SC2 field-valid definition: “
- Endogenous genetic deletion, suppression, or degron depletion of the proposed mediator in vivo
- Selective pathway inhibition that changes onset/progression while preserving broader viability
- Vagotomy-type natural experiment, only if confounding is rigorously controlled AND effect is disease-initiation relevant rather than merely associated” Positive full-scope result: yes. Endogenous cell-lineage perturbations of mutant SOD1, NFκB and astrocytic APOE, as well as complementary immune manipulations, materially alter survival, atrophy or function in specified models. Defined mediator causality crosses Δ and merits Ε with explicit genotype, state and timing restrictions. Protective SYK, TREM2, IL3 and innate signaling states rule out indiscriminate activation suppression.

SC3 field-valid definition: “
- Endogenous genetic deletion, suppression, or degron depletion of the proposed mediator in vivo
- Selective pathway inhibition that changes onset/progression while preserving broader viability
- Vagotomy-type natural experiment, only if confounding is rigorously controlled AND effect is disease-initiation relevant rather than merely associated” Positive full-scope result: yes. Glial contributions exceed a passive bystander or clearance-only account. SOD1 removal in glia and partial NFκB suppression slow later progression without changing onset; these are reactive causal amplification, which qualifies under the registered wording. Astrocyte and microglial effects are not interchangeable and do not establish obligatory upstream initiation. This defined causal contribution is Ε.

SC4 field-valid definition: “
- Endogenous genetic deletion, suppression, or degron depletion of the proposed mediator in vivo
- Selective pathway inhibition that changes onset/progression while preserving broader viability
- Vagotomy-type natural experiment, only if confounding is rigorously controlled AND effect is disease-initiation relevant rather than merely associated” Positive full-scope result: no. AD, PD and ALS models show context-dependent causal immune effects, but human pathway engagement has not reliably produced sustained clinical benefit. AL002, NLY01, AZA-PD, ADAPT, NSAID treatment and minocycline test different, incompletely selective mechanisms. Protective and harmful states require prospective boundaries; the general disease-class claim remains Γ.

### Upward threshold ladder

SC1 Δ: “Requires at least one genuine causal perturbation in a disease-relevant in vivo system, or equivalent human natural experiment that materially supports the mechanism.” Satisfied: no. Prospective, state-resolved human measurements must precede independently measured injury without relying solely on terminal tissue or inferred pseudotime.

SC2 Δ: “Requires at least one genuine causal perturbation in a disease-relevant in vivo system, or equivalent human natural experiment that materially supports the mechanism.” Satisfied: yes. 

SC2 Ε: “Requires causal support in a defined context, but with clear restriction by subtype, disease stage, model, or tissue.

Use Ε when there is a real causal contribution, but generalization to the entire disease class fails.” Satisfied: yes. 

SC2 Ζ: “Requires:
- Multiple independent causal lines in vivo
- Replication across at least two relevant systems
- Narrowing of major alternative models” Satisfied: no. Independent mediator loss and matched rescue after onset should alter neuron survival and function while retaining normal clearance, peripheral immunity and baseline viability.

SC3 Δ: “Requires at least one genuine causal perturbation in a disease-relevant in vivo system, or equivalent human natural experiment that materially supports the mechanism.” Satisfied: yes. 

SC3 Ε: “Requires causal support in a defined context, but with clear restriction by subtype, disease stage, model, or tissue.

Use Ε when there is a real causal contribution, but generalization to the entire disease class fails.” Satisfied: yes. 

SC3 Ζ: “Requires:
- Multiple independent causal lines in vivo
- Replication across at least two relevant systems
- Narrowing of major alternative models” Satisfied: no. Bidirectional, stage-controlled mediator experiments should distinguish causal amplification from passive response while matching the initiating injury; initiation is not required for amplification.

SC4 Δ: “Requires at least one genuine causal perturbation in a disease-relevant in vivo system, or equivalent human natural experiment that materially supports the mechanism.” Satisfied: no. Prespecified human cell-state target engagement and sustained clinical trajectory changes, together with independent causal replications in AD, PD and ALS, would cross the generalization gate.

Upper ladder: TREM2/PLCG2/ABI3 risk genetics and randomized clinical immune interventions were considered. Genetics has directional and functional heterogeneity; AL002 has CSF engagement but a clinical null, and broad drugs have uncertain glial specificity. These data do not establish a universal mediator-to-clinical trajectory chain. Higher grades remain available to prospectively defined states with convergent causal human and model evidence.

Independence: The 69 reports are not 69 independent causal lines. ADAPT follow-up reuses the original randomization. Several SOD1, P301S/APOE and 5xFAD studies share engineered backgrounds and laboratories. Genetic cohorts can overlap. Orthogonal timing, cell-lineage, protection and clinical endpoint contrasts cover different weaknesses, but none yet settles the broad human state-specific claim.

Cell-lineage and timing perturbations [49, 48, 52, 25]: Exclude a wholly passive glial role in defined models; shared proteinopathy backgrounds and pleiotropy remain.

Protective-state loss and rescue [14, 2, 51]: Exclude indiscriminate harmful activation and expose clearance/state dependence.

Adaptive immune hub perturbations [68, 5]: Exclude a solely neuron-autonomous progression account, while retaining peripheral and aggregate-burden mediation.

Randomized human interventions [40, 42, 43, 39, 60, 61]: Narrow treatment confounding but do not universally verify cell-state engagement; null, harmful and mixed results are retained.

Human risk genetics [55, 56, 57]: Support immune relevance, but risk alleles do not identify a single inflammatory mediator or prove suppression should help.

### Falsification and pitfalls

Activation can protect as well as injure. SYK/TREM2/IL3 and TRIF/MyD88 loss can worsen pathology or outcome; reduced activation is not uniformly protective. Additional downgrade: False.

Causal amplification need not prevent onset. Boillee, Yamanaka, Frakes and Guttenplan change later progression with little or no onset effect. Additional downgrade: False.

Clinical engagement does not guarantee trajectory benefit. AL002 CSF engagement accompanies a clinical null; minocycline worsens ALSFRS-R decline; short sargramostim signals are mixed. Additional downgrade: False.

Reactive and causal treated as mutually exclusive. Original causal-driver wording and SOD1 timing experiments.

Activation treated as one uniform harmful state. TREM2/SYK/IL3 and protective immune programs.

Provider retrieval failure treated as missing clinical literature. Consensus did not retrieve named trials; original PubMed reports were examined.

Short MMSE signal presented as established benefit. Sargramostim baseline imbalance, ADAS-Cog13 worsening and null functional endpoints.

Historical score presented as blinded derivation. Historical tau/neuro Γ levels were visible before new adjudication; explicitly unblinded.

### Adversarial steelman: above, then below

The strongest Δ/Ε case assembles cell-lineage SOD1, partial NFκB, astrocytic APOE4 and cytokine deletion with neuronal survival and functional outcomes, plus human risk genetics. It convincingly crosses causal contribution in defined contexts. It does not establish one harmful activation exposure throughout AD, PD and ALS or sustained selective human clinical mediation.

The strongest Β case treats all activation as correlated response to injured neurons. That fails against progression-changing lineage perturbations and matched controls. Reactive causal amplification is experimentally supported and deserves Ε for the defined subclaims; Γ for the full conjunction reflects the unresolved generalization gate.

Reject a passive-bystander-only account, retain defined causal amplification, and reserve stronger disease-class grades for prospective state-specific generalization. Do not lower the score merely because the glial response follows an initiating injury.

### Final score, causal direction and M/S/D

The registered weakest-link conjunction returns Γ because disease-class human generalization remains unresolved. Defined glial states have real causal effects on neuronal loss and progression, including reactive amplifiers. This evidence is not dismissed for being downstream. Opposite effects of protective immune programs and negative or harmful human interventions prevent treating chronic activation as one broadly pathogenic exposure. The limiting gate is generalization and state-specific human mediation, not an added obligation to initiate disease.

Defined glial and immune states causally amplify established disease. The original neuronal lesion can precede their harmful contribution, so reactive origin does not imply noncausality. Protective states can oppose injury. This conditional amplification is supported more directly than a general initiating or rate-limiting role.

Momentum: Mechanistic only. Completed 2026 AL002 and AZA-PD reports and recent state-selective models refine the clinical and protective-state boundary. No pending decisive result is claimed.

Stability: Stable. No verified pending report is asserted as likely to change the registered broad level imminently.

Divergence: Contested; no single implied-level arithmetic. These dimensions do not change level or confidence.

## PROPOSED DECISIVE EXPERIMENTS

Induce loss of a prespecified harmful glial program after comparable neuronal pathology begins, preserve protective clearance and peripheral immune function, then restore the same mediator with cell-specific rescue in independent AD, PD and ALS settings.

Readout: Blinded neuron counts, functional decline, survival where relevant, and direct cell-state/target engagement. Controls: Matched initiating protein burden, stage, neuronal viability, clearance, peripheral immunity and rescue expression. Supporting: Post-onset suppression slows loss and functional decline; matched rescue restores both with replicated prespecified state boundaries. Falsifying: Verified selective loss leaves degeneration unchanged, or protection is explained wholly by altered initial protein burden or nonspecific immune effects.

## RESEARCH TIMELINE

| Year | Milestone | Primary source | Interpretation |
|---|---|---|---|
| 2003 | Symptomatic AD NSAID trial null | [PMID12783912](https://pubmed.ncbi.nlm.nih.gov/12783912/) | Clinical strategy limitation. |
| 2006 | Neuron and microglial SOD1 timing separates onset/progression | [PMID16741123](https://pubmed.ncbi.nlm.nih.gov/16741123/) | Reactive contribution can be causal. |
| 2007 | ADAPT early prevention result | [PMID17460158](https://pubmed.ncbi.nlm.nih.gov/17460158/) | Same cohort as later follow-up. |
| 2007 | Minocycline accelerates ALS functional decline | [PMID17980667](https://pubmed.ncbi.nlm.nih.gov/17980667/) | Harmful clinical contrast. |
| 2008 | Astrocytic mutant SOD1 controls later progression | [PMID18246065](https://pubmed.ncbi.nlm.nih.gov/18246065/) | No obligatory onset effect. |
| 2013 | Human TREM2 risk variants | [PMID23150934](https://pubmed.ncbi.nlm.nih.gov/23150934/) | Immune genetics, heterogeneous function. |
| 2013 | ADAPT extended follow-up | [PMID23562431](https://pubmed.ncbi.nlm.nih.gov/23562431/) | No overall prevention. |
| 2014 | Partial myeloid NFκB inhibition extends model survival | [PMID24607225](https://pubmed.ncbi.nlm.nih.gov/24607225/) | Lineage and systemic limits. |
| 2016 | Complement-mediated early synapse loss | [PMID27033548](https://pubmed.ncbi.nlm.nih.gov/27033548/) | AD model contribution. |
| 2017 | Microglial induction of toxic astrocytes | [PMID28099414](https://pubmed.ncbi.nlm.nih.gov/28099414/) | Defined state, not all reactivity. |
| 2017 | TREM2 loss protects a pure-tau model | [PMID29073081](https://pubmed.ncbi.nlm.nih.gov/29073081/) | Context-dependent sign. |
| 2018 | NLRP3 inhibition protects multiple PD models | [PMID30381407](https://pubmed.ncbi.nlm.nih.gov/30381407/) | Model trajectory contribution. |
| 2019 | Microglial clearance failure worsens AD model | [PMID31257024](https://pubmed.ncbi.nlm.nih.gov/31257024/) | Protective function matters. |
| 2020 | Triple cytokine deletion delays ALS progression | [PMID32719333](https://pubmed.ncbi.nlm.nih.gov/32719333/) | Delayed rather than prevented loss. |
| 2021 | Astrocytic APOE4 deletion protects established tau model | [PMID33831349](https://pubmed.ncbi.nlm.nih.gov/33831349/) | Genotype and timing boundary. |
| 2021 | Sargramostim AD safety study | [PMID33778150](https://pubmed.ncbi.nlm.nih.gov/33778150/) | Mixed brief cognitive findings. |
| 2022 | SYK supports protective plaque-associated microglia | [PMID36306735](https://pubmed.ncbi.nlm.nih.gov/36306735/) | Reduced activation can worsen injury. |
| 2023 | Microglia–T-cell hubs modify tau degeneration | [PMID36890231](https://pubmed.ncbi.nlm.nih.gov/36890231/) | Peripheral and CNS interaction. |
| 2024 | NLY01 PD efficacy trial null | [PMID38101901](https://pubmed.ncbi.nlm.nih.gov/38101901/) | Exploratory age signal not definitive. |
| 2025 | NT0796 short open-label CNS engagement | [PMID40792655](https://pubmed.ncbi.nlm.nih.gov/40792655/) | Biomarkers without established efficacy. |
| 2026 | AL002 TREM2 agonist clinical null | [PMID41787076](https://pubmed.ncbi.nlm.nih.gov/41787076/) | CSF engagement is insufficient. |
| 2026 | AZA-PD primary outcome null | [PMID41389828](https://pubmed.ncbi.nlm.nih.gov/41389828/) | Peripheral immune strategy boundary. |

## KNOWLEDGE GAPS

Define harmful and protective microglial/astrocytic states prospectively rather than assigning labels after outcome.

Use post-onset selective loss and matched rescue to separate initiation, amplification and recovery.

Match initiating protein burden and intrinsic neuronal toxicity so aggregate lowering does not masquerade as glial mediation.

Verify human cell-state engagement and sustained functional outcomes, preserving null and harmful results alongside positive biomarkers.

Map amyloid, tau, APOE, TREM2 and disease-stage boundaries of protective clearance before targeting activation broadly.

Resolve peripheral adaptive-immune contributions and systemic drug effects separately from CNS glial mediation.

## METHODS AND CONSENSUS CONTRIBUTION

Consensus was used through the signed-in browser for discovery, counterevidence, research-gap and timeline suggestions, and attempted checks of specific indexed full-text passages. Returned Methods/Results passages were reconciled with original-paper passages where available; failed retrievals, source substitutions and unverified aspects are recorded in the case reconciliation. Successful passage checks are distinguished from provider summaries. Provider screening counts, citation counts and meters were not used as primary counts or HSEF scores. This source log records access depth; it does not claim complete-paper or supplement review. The monthly API allowance was conserved: zero Consensus API calls in this batch.

## REVIEWER NOTES

69 primary reports; six selected original-body examinations, remaining abstracts only. No complete-paper or supplement audit. Original source content and linked-status metadata were examined. Some correction details remain unexamined and are flagged in cards. Consensus two Pro browser queries, zero Deep and zero API; no claim of provider completeness.

Search leads, reviews, protocols and correction notices do not pad the primary count. Downloaded XML does not earn examination credit. Linked notices remain attached; unexamined correction details limit inference. This is a coordinator source/claim consistency assessment, not independent blinded calibration.

## ADVERSARIAL REVIEW AND APPROVAL

- **Authoring model:** Sol 6.1. **Adversarial reviewer:** Claude Opus 5.5 (Anthropic). **Date:** 2026-10-06.
- **Scope of checks:** PMIDs verified 69/69 against PubMed (title, first author, year, journal; all resolve and match; no retraction found; the linked erratum on [68] is already flagged). Load-bearing claims spot-checked: 19 (abstracts via PubMed; local original-body passages for [PMID24607225], [PMID32719333], [PMID33778150]). Timeline years checked against cited records (22/22 consistent; e-pub/print ±1 accepted). Scoring logic checked against SKILL.md Step 3, hsef-framework.md hard constraints and cap precedence, rubric 01 Δ/Ε thresholds and rubric 05 Γ hard rules.
- **Scoring result:** unchanged. hypothesis-index.md prescribes weakest-link for neuroinflammation-first models; weakest-link over SC1 Γ, SC2 Ε, SC3 Ε, SC4 Γ returns **Γ** (primary-subclaim and scope-matched alternatives Ε are correctly reported as sensitivities only). SC2/SC3 Ε rest on endogenous cell-lineage or pathway-specific in vivo perturbations (Cre-excision of mutant SOD1 in microglia/astrocytes, CSF1R-cre heterozygous IKKβ deletion, inducible astrocytic APOE4 removal, IL1α/TNF/C1q deletion) with explicit genotype/stage/model restriction, which meets the rubric 01 Ε text ("causal support in a defined context, but with clear restriction"). The transgenic SOD1/P301S backgrounds are the disease system, not the mediator perturbation, so the overexpression cap binds only the inference to sporadic human disease (cap precedence), which SC4 already carries at Γ.

| ID | Severity | Location | Original → corrected | Evidence |
|---|---|---|---|---|
| AR-1 | Editorial | REFERENCES 58; scorecard citation strings | "58.  , Lyketsos CG, Breitner JC et al." → "58. ADAPT Research Group, Lyketsos CG, Breitner JC et al." | PubMed record PMID 17460158 lists collective author "ADAPT Research Group" first (builder dropped collective name). |
| AR-2 | Editorial | REFERENCES 59; scorecard citation strings | "59.   et al. Results of a follow-up…" → "59. Alzheimer's Disease Anti-inflammatory Prevention Trial Research Group. Results of a follow-up…" | PubMed record PMID 23562431 has a sole collective author with that name. |
| AR-3 | Editorial | references.ris (records for PMIDs 17460158, 23562431, 17980667, 23150934, 12783912, 28714976) | 7 blank "AU  - " lines → collective names (ADAPT Research Group; Alzheimer's Disease Anti-inflammatory Prevention Trial Research Group; Western ALS Study Group; Alzheimer Genetic Analysis Group; Alzheimer's Disease Cooperative Study; ARUK Consortium; GERAD/PERADES, CHARGE, ADGC, EADI) | PubMed efetch CollectiveName fields at the matching author positions. |

**Spot-checks confirmed (no change):** minocycline ALS faster ALSFRS-R decline with non-significant mortality HR 1.32 [39]; AL002 CSF target engagement, primary CDR-SB null, ARIA-like events most frequent [40]; NLY01 primary null with exploratory younger-age signal [42]; AZA-PD primary null [43]; evoke/evoke+ semaglutide null [41]; rofecoxib/naproxen symptomatic AD null [60]; ADAPT early (non-significant HRs >1) and ADAPT-FS (no overall prevention, same cohort) [58, 59]; Boillée microglial SOD1 removal slows later progression not onset [49]; Yamanaka astrocytic SOD1 removal slows progression not onset [48]; Frakes heterozygous CSF1R-cre IKKβ deletion: onset unchanged, median survival +20 days, mutant SOD1 levels unchanged, homozygotes immune-compromised, CSF1R-cre also peripheral myeloid [52]; Guttenplan triple knockout preserves motor neurons at control endpoint with similar loss at later endpoint [PMID32719333]; sargramostim AD baseline MMSE imbalance (17.1 vs 20.75) and ADAS-Cog13 worsening at follow-up [PMID33778150]; SYK deficiency worsens pathology/behaviour and anti-CLEC7A rescues microglial activation [14]; TREM2 loss protects PS19 tau model [69]; LANDO-deficient microglia accelerate AD-model degeneration (timeline 2019); NT-0796 open-label 28-day CSF engagement in 10 PD subjects; TREM2/PLCG2/ABI3 genetics [55–57].

**Rejected findings (checked, no change):** (1) The case's subclaims replace the index's SC2 "inflammation upstream of proteinopathy" with SC3 "beyond a purely reactive response", and SC3 credits reactive amplification as primacy. This is a stated, preregistered judgement tied to the registered wording; scoring the index primacy subclaim could not raise the headline and, at worst Γ, would leave weakest-link at Γ. Retained as a residual limitation, not changed. (2) "ADAPT … trials are null" although early ADAPT HRs pointed toward harm (non-significant): accurate as stated. (3) Year differences of ±1 between e-pub and issue (e.g., TREM2 NEJM 2012/2013): accepted by convention.

**Residual limitations:** No complete-paper audit; only 6 partial bodies examined by the author and 3 re-read here. The SC3 interpretation of "rather than reactive" as "not purely passive" is a defensible but contestable reading; readers wanting the inflammation-first (initiation) claim should treat that claim as not graded above Γ here. Batch-level files (READER.html, SOURCE_LEDGER.csv if it carries author strings) still contain the uncorrected reference strings and must be propagated by the lead.

**Approval:** Following adversarial review and the corrections listed above, the revised document is approved. — Claude Opus 5.5 (Anthropic), 2026-10-06.

## REFERENCES

1. Smajić S, Prada-Medina CA, Landoulsi Z et al. Single-cell sequencing of human midbrain reveals glial activation and a Parkinson-specific neuronal state. Brain : a journal of neurology. 2022. [PMID 34919646](https://pubmed.ncbi.nlm.nih.gov/34919646/). DOI: 10.1093/brain/awab446. Access: abstract_metadata. Study: PMID34919646.

2. McAlpine CS, Park J, Griciuc A et al. Astrocytic interleukin-3 programs microglia and limits Alzheimer's disease. Nature. 2021. [PMID 34262178](https://pubmed.ncbi.nlm.nih.gov/34262178/). DOI: 10.1038/s41586-021-03734-6. Access: abstract_metadata. Study: PMID34262178.

3. Gabitto MI, Travaglini KJ, Rachleff VM et al. Integrated multimodal cell atlas of Alzheimer's disease. Nature neuroscience. 2024. [PMID 39402379](https://pubmed.ncbi.nlm.nih.gov/39402379/). DOI: 10.1038/s41593-024-01774-5. Access: abstract_metadata. Study: PMID39402379.

4. Yan YQ, Zheng R, Liu Y et al. Parkin regulates microglial NLRP3 and represses neurodegeneration in Parkinson's disease. Aging cell. 2023. [PMID 37029500](https://pubmed.ncbi.nlm.nih.gov/37029500/). DOI: 10.1111/acel.13834. Access: abstract_metadata. Study: PMID37029500.

5. Bido S, Nannoni M, Muggeo S et al. Microglia-specific IL-10 gene delivery inhibits neuroinflammation and neurodegeneration in a mouse model of Parkinson's disease. Science translational medicine. 2024. [PMID 39167665](https://pubmed.ncbi.nlm.nih.gov/39167665/). DOI: 10.1126/scitranslmed.adm8563. Access: abstract_metadata. Study: PMID39167665.

6. Yun SP, Kam TI, Panicker N et al. Block of A1 astrocyte conversion by microglia is neuroprotective in models of Parkinson's disease. Nature medicine. 2018. [PMID 29892066](https://pubmed.ncbi.nlm.nih.gov/29892066/). DOI: 10.1038/s41591-018-0051-5. Access: abstract_metadata. Study: PMID29892066.

7. Gordon R, Albornoz EA, Christie DC et al. Inflammasome inhibition prevents α-synuclein pathology and dopaminergic neurodegeneration in mice. Science translational medicine. 2018. [PMID 30381407](https://pubmed.ncbi.nlm.nih.gov/30381407/). DOI: 10.1126/scitranslmed.aah4066. Access: abstract_metadata. Study: PMID30381407.

8. Dejanovic B, Wu T, Tsai MC et al. Complement C1q-dependent excitatory and inhibitory synapse elimination by astrocytes and microglia in Alzheimer's disease mouse models. Nature aging. 2022. [PMID 37118504](https://pubmed.ncbi.nlm.nih.gov/37118504/). DOI: 10.1038/s43587-022-00281-1. Access: abstract_metadata. Study: PMID37118504.

9. Qin L, Wu X, Block ML et al. Systemic LPS causes chronic neuroinflammation and progressive neurodegeneration. Glia. 2007. [PMID 17203472](https://pubmed.ncbi.nlm.nih.gov/17203472/). DOI: 10.1002/glia.20467. Access: abstract_metadata. Study: PMID17203472.

10. Tsunemi T, Ishiguro Y, Yoroisaka A et al. Astrocytes Protect Human Dopaminergic Neurons from α-Synuclein Accumulation and Propagation. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2020. [PMID 33046546](https://pubmed.ncbi.nlm.nih.gov/33046546/). DOI: 10.1523/JNEUROSCI.0954-20.2020. Access: abstract_metadata. Study: PMID33046546.

11. Castro-Sánchez S, García-Yagüe ÁJ, López-Royo T et al. Cx3cr1-deficiency exacerbates alpha-synuclein-A53T induced neuroinflammation and neurodegeneration in a mouse model of Parkinson's disease. Glia. 2018. [PMID 29624735](https://pubmed.ncbi.nlm.nih.gov/29624735/). DOI: 10.1002/glia.23338. Access: abstract_metadata. Study: PMID29624735.

12. Hong S, Beja-Glasser VF, Nfonoyim BM et al. Complement and microglia mediate early synapse loss in Alzheimer mouse models. Science (New York, N.Y.). 2016. [PMID 27033548](https://pubmed.ncbi.nlm.nih.gov/27033548/). DOI: 10.1126/science.aad8373. Access: abstract_metadata. Study: PMID27033548.

13. Ulland TK, Song WM, Huang SC et al. TREM2 Maintains Microglial Metabolic Fitness in Alzheimer's Disease. Cell. 2017. [PMID 28802038](https://pubmed.ncbi.nlm.nih.gov/28802038/). DOI: 10.1016/j.cell.2017.07.023. Access: abstract_metadata. Study: PMID28802038.

14. Wang S, Sudan R, Peng V et al. TREM2 drives microglia response to amyloid-β via SYK-dependent and -independent pathways. Cell. 2022. [PMID 36306735](https://pubmed.ncbi.nlm.nih.gov/36306735/). DOI: 10.1016/j.cell.2022.09.033. Access: full_text_partial. Study: PMID36306735.

15. Zhong L, Sheng X, Wang W et al. TREM2 receptor protects against complement-mediated synaptic loss by binding to complement C1q during neurodegeneration. Immunity. 2023. [PMID 37442133](https://pubmed.ncbi.nlm.nih.gov/37442133/). DOI: 10.1016/j.immuni.2023.06.016. Access: abstract_metadata. Study: PMID37442133.

16. Zhou J, Wade SD, Graykowski D et al. The neuronal pentraxin Nptx2 regulates complement activity and restrains microglia-mediated synapse loss in neurodegeneration. Science translational medicine. 2023. [PMID 36989373](https://pubmed.ncbi.nlm.nih.gov/36989373/). DOI: 10.1126/scitranslmed.adf0141. Access: abstract_metadata. Study: PMID36989373.

17. Wang Y, Cella M, Mallinson K et al. TREM2 lipid sensing sustains the microglial response in an Alzheimer's disease model. Cell. 2015. [PMID 25728668](https://pubmed.ncbi.nlm.nih.gov/25728668/). DOI: 10.1016/j.cell.2015.01.049. Access: abstract_metadata. Study: PMID25728668.

18. Heckmann BL, Teubner BJW, Tummers B et al. LC3-Associated Endocytosis Facilitates β-Amyloid Clearance and Mitigates Neurodegeneration in Murine Alzheimer's Disease. Cell. 2019. [PMID 31257024](https://pubmed.ncbi.nlm.nih.gov/31257024/). DOI: 10.1016/j.cell.2019.05.056. Access: abstract_metadata. Study: PMID31257024.

19. Gratuze M, Leyns CE, Sauerbeck AD et al. Impact of TREM2R47H variant on tau pathology-induced gliosis and neurodegeneration. The Journal of clinical investigation. 2020. [PMID 32544086](https://pubmed.ncbi.nlm.nih.gov/32544086/). DOI: 10.1172/JCI138179. Access: abstract_metadata. Study: PMID32544086.

20. Clarke N, Thornton P, Reader V et al. Anti-Neuroinflammatory and Anti-Inflammatory Effects of the NLRP3 Inhibitor NT-0796 in Subjects with Parkinson's Disease. Movement disorders : official journal of the Movement Disorder Society. 2025. [PMID 40792655](https://pubmed.ncbi.nlm.nih.gov/40792655/). DOI: 10.1002/mds.30307. Access: abstract_metadata. Study: PMID40792655.

21. Liang Y, Li S, Guo Q et al. Complement 3-deficient mice are not protected against MPTP-induced dopaminergic neurotoxicity. Brain research. 2007. [PMID 17900537](https://pubmed.ncbi.nlm.nih.gov/17900537/). DOI: 10.1016/j.brainres.2007.08.033. Access: abstract_metadata. Study: PMID17900537.

22. Carpanini SM, Torvell M, Bevan RJ et al. Terminal complement pathway activation drives synaptic loss in Alzheimer's disease models. Acta neuropathologica communications. 2022. [PMID 35794654](https://pubmed.ncbi.nlm.nih.gov/35794654/). DOI: 10.1186/s40478-022-01404-w. Access: abstract_metadata. Study: PMID35794654.

23. Bido S, Muggeo S, Massimino L et al. Microglia-specific overexpression of α-synuclein leads to severe dopaminergic neurodegeneration by phagocytic exhaustion and oxidative toxicity. Nature communications. 2021. [PMID 34716339](https://pubmed.ncbi.nlm.nih.gov/34716339/). DOI: 10.1038/s41467-021-26519-x. Access: abstract_metadata. Study: PMID34716339.

24. Hinkle JT, Patel J, Panicker N et al. STING mediates neurodegeneration and neuroinflammation in nigrostriatal α-synucleinopathy. Proceedings of the National Academy of Sciences of the United States of America. 2022. [PMID 35394877](https://pubmed.ncbi.nlm.nih.gov/35394877/). DOI: 10.1073/pnas.2118819119. Access: abstract_metadata. Study: PMID35394877.

25. Wang C, Xiong M, Gratuze M et al. Selective removal of astrocytic APOE4 strongly protects against tau-mediated neurodegeneration and decreases synaptic phagocytosis by microglia. Neuron. 2021. [PMID 33831349](https://pubmed.ncbi.nlm.nih.gov/33831349/). DOI: 10.1016/j.neuron.2021.03.024. Access: full_text_partial. Study: PMID33831349.

26. Gomez-Arboledas A, Davila JC, Sanchez-Mejias E et al. Phagocytic clearance of presynaptic dystrophies by reactive astrocytes in Alzheimer's disease. Glia. 2018. [PMID 29178139](https://pubmed.ncbi.nlm.nih.gov/29178139/). DOI: 10.1002/glia.23270. Access: abstract_metadata. Study: PMID29178139.

27. Han X, Li Q, Lan X et al. Microglial Depletion with Clodronate Liposomes Increases Proinflammatory Cytokine Levels, Induces Astrocyte Activation, and Damages Blood Vessel Integrity. Molecular neurobiology. 2019. [PMID 30734229](https://pubmed.ncbi.nlm.nih.gov/30734229/). DOI: 10.1007/s12035-019-1502-9. Access: abstract_metadata. Study: PMID30734229.

28. Liddelow SA, Guttenplan KA, Clarke LE et al. Neurotoxic reactive astrocytes are induced by activated microglia. Nature. 2017. [PMID 28099414](https://pubmed.ncbi.nlm.nih.gov/28099414/). DOI: 10.1038/nature21029. Access: abstract_metadata. Study: PMID28099414.

29. Keren-Shaul H, Spinrad A, Weiner A et al. A Unique Microglia Type Associated with Restricting Development of Alzheimer's Disease. Cell. 2017. [PMID 28602351](https://pubmed.ncbi.nlm.nih.gov/28602351/). DOI: 10.1016/j.cell.2017.05.018. Access: abstract_metadata. Study: PMID28602351.

30. Scheiblich H, Eikens F, Wischhof L et al. Microglia rescue neurons from aggregate-induced neuronal dysfunction and death through tunneling nanotubes. Neuron. 2024. [PMID 39059388](https://pubmed.ncbi.nlm.nih.gov/39059388/). DOI: 10.1016/j.neuron.2024.06.029. Access: abstract_metadata. Study: PMID39059388.

31. Ising C, Venegas C, Zhang S et al. NLRP3 inflammasome activation drives tau pathology. Nature. 2019. [PMID 31748742](https://pubmed.ncbi.nlm.nih.gov/31748742/). DOI: 10.1038/s41586-019-1769-z. Access: abstract_metadata. Study: PMID31748742.

32. Fernández-Matarrubia M, Valera-Barrero A, Renuncio-García M et al. Early microglial and astrocyte reactivity in preclinical Alzheimer's disease. Alzheimer's & dementia : the journal of the Alzheimer's Association. 2025. [PMID 40747577](https://pubmed.ncbi.nlm.nih.gov/40747577/). DOI: 10.1002/alz.70502. Access: abstract_metadata. Study: PMID40747577.

33. Kummer MP, Ising C, Kummer C et al. Microglial PD-1 stimulation by astrocytic PD-L1 suppresses neuroinflammation and Alzheimer's disease pathology. The EMBO journal. 2021. [PMID 34825707](https://pubmed.ncbi.nlm.nih.gov/34825707/). DOI: 10.15252/embj.2021108662. Access: abstract_metadata. Study: PMID34825707.

34. Venegas C, Kumar S, Franklin BS et al. Microglia-derived ASC specks cross-seed amyloid-β in Alzheimer's disease. Nature. 2017. [PMID 29293211](https://pubmed.ncbi.nlm.nih.gov/29293211/). DOI: 10.1038/nature25158. Access: abstract_metadata. Study: PMID29293211.

35. Wang S, Mustafa M, Yuede CM et al. Anti-human TREM2 induces microglia proliferation and reduces pathology in an Alzheimer's disease model. The Journal of experimental medicine. 2020. [PMID 32579671](https://pubmed.ncbi.nlm.nih.gov/32579671/). DOI: 10.1084/jem.20200785. Access: abstract_metadata. Study: PMID32579671.

36. Gratuze M, Schlachetzki JCM, D'Oliveira Albanus R et al. TREM2-independent microgliosis promotes tau-mediated neurodegeneration in the presence of ApoE4. Neuron. 2023. [PMID 36368315](https://pubmed.ncbi.nlm.nih.gov/36368315/). DOI: 10.1016/j.neuron.2022.10.022. Access: abstract_metadata. Study: PMID36368315.

37. Long H, Simmons A, Mayorga A et al. Preclinical and first-in-human evaluation of AL002, a novel TREM2 agonistic antibody for Alzheimer's disease. Alzheimer's research & therapy. 2024. [PMID 39444037](https://pubmed.ncbi.nlm.nih.gov/39444037/). DOI: 10.1186/s13195-024-01599-1. Access: abstract_metadata. Study: PMID39444037.

38. Takalo M, Wittrahm R, Wefers B et al. The Alzheimer's disease-associated protective Plcγ2-P522R variant promotes immune functions. Molecular neurodegeneration. 2020. [PMID 32917267](https://pubmed.ncbi.nlm.nih.gov/32917267/). DOI: 10.1186/s13024-020-00402-7. Access: abstract_metadata. Study: PMID32917267.

39. Gordon PH, Moore DH, Miller RG et al. Efficacy of minocycline in patients with amyotrophic lateral sclerosis: a phase III randomised trial. The Lancet. Neurology. 2007. [PMID 17980667](https://pubmed.ncbi.nlm.nih.gov/17980667/). DOI: 10.1016/S1474-4422(07)70270-3. Access: abstract_metadata. Study: PMID17980667.

40. Mummery CJ, Mayorga AJ, Simmons A et al. The TREM2 agonistic antibody AL002 in early Alzheimer's disease: a phase 2 randomized trial. Nature medicine. 2026. [PMID 41787076](https://pubmed.ncbi.nlm.nih.gov/41787076/). DOI: 10.1038/s41591-026-04273-1. Access: abstract_metadata. Study: PMID41787076.

41. Cummings JL, Atri A, Sano M et al. Efficacy and safety of oral semaglutide 14 mg (flexible dose) in early-stage symptomatic Alzheimer's disease (evoke and evoke+): two phase 3, randomised, placebo-controlled trials. Lancet (London, England). 2026. [PMID 41865758](https://pubmed.ncbi.nlm.nih.gov/41865758/). DOI: 10.1016/S0140-6736(26)00459-9. Access: abstract_metadata. Study: EVOKE-and-EVOKEplus-two-trials-one-report.

42. McGarry A, Rosanbalm S, Leinonen M et al. Safety, tolerability, and efficacy of NLY01 in early untreated Parkinson's disease: a randomised, double-blind, placebo-controlled trial. The Lancet. Neurology. 2024. [PMID 38101901](https://pubmed.ncbi.nlm.nih.gov/38101901/). DOI: 10.1016/S1474-4422(23)00378-2. Access: abstract_metadata. Study: PMID38101901.

43. Greenland JC, Dresser K, Cutting E et al. Azathioprine for the treatment of early Parkinson's disease (AZA-PD): a randomised, double-blind, placebo-controlled, proof-of-concept, phase 2 trial. The Lancet. Neurology. 2026. [PMID 41389828](https://pubmed.ncbi.nlm.nih.gov/41389828/). DOI: 10.1016/S1474-4422(25)00386-2. Access: abstract_metadata. Study: PMID41389828.

44. Gendelman HE, Zhang Y, Santamaria P et al. Evaluation of the safety and immunomodulatory effects of sargramostim in a randomized, double-blind phase 1 clinical Parkinson's disease trial. NPJ Parkinson's disease. 2017. [PMID 28649610](https://pubmed.ncbi.nlm.nih.gov/28649610/). DOI: 10.1038/s41531-017-0013-5. Access: abstract_metadata. Study: PMID28649610.

45. Brenner D, Sieverding K, Bruno C et al. Heterozygous Tbk1 loss has opposing effects in early and late stages of ALS in mice. The Journal of experimental medicine. 2019. [PMID 30635357](https://pubmed.ncbi.nlm.nih.gov/30635357/). DOI: 10.1084/jem.20180729. Access: abstract_metadata. Study: PMID30635357.

46. Mesci P, Zaïdi S, Lobsiger CS et al. System xC- is a mediator of microglial function and its deletion slows symptoms in amyotrophic lateral sclerosis mice. Brain : a journal of neurology. 2015. [PMID 25384799](https://pubmed.ncbi.nlm.nih.gov/25384799/). DOI: 10.1093/brain/awu312. Access: abstract_metadata. Study: PMID25384799.

47. Lobsiger CS, Boillée S, Pozniak C et al. C1q induction and global complement pathway activation do not contribute to ALS toxicity in mutant SOD1 mice. Proceedings of the National Academy of Sciences of the United States of America. 2013. [PMID 24170856](https://pubmed.ncbi.nlm.nih.gov/24170856/). DOI: 10.1073/pnas.1318309110. Access: abstract_metadata. Study: PMID24170856.

48. Yamanaka K, Chun SJ, Boillee S et al. Astrocytes as determinants of disease progression in inherited amyotrophic lateral sclerosis. Nature neuroscience. 2008. [PMID 18246065](https://pubmed.ncbi.nlm.nih.gov/18246065/). DOI: 10.1038/nn2047. Access: abstract_metadata. Study: PMID18246065.

49. Boillée S, Yamanaka K, Lobsiger CS et al. Onset and progression in inherited ALS determined by motor neurons and microglia. Science (New York, N.Y.). 2006. [PMID 16741123](https://pubmed.ncbi.nlm.nih.gov/16741123/). DOI: 10.1126/science.1123511. Access: abstract_metadata. Study: PMID16741123.

50. Endo F, Komine O, Fujimori-Tonou N et al. Astrocyte-derived TGF-β1 accelerates disease progression in ALS mice by interfering with the neuroprotective functions of microglia and T cells. Cell reports. 2015. [PMID 25892237](https://pubmed.ncbi.nlm.nih.gov/25892237/). DOI: 10.1016/j.celrep.2015.03.053. Access: abstract_metadata. Study: PMID25892237.

51. Komine O, Yamashita H, Fujimori-Tonou N et al. Innate immune adaptor TRIF deficiency accelerates disease progression of ALS mice with accumulation of aberrantly activated astrocytes. Cell death and differentiation. 2018. [PMID 29568058](https://pubmed.ncbi.nlm.nih.gov/29568058/). DOI: 10.1038/s41418-018-0098-3. Access: abstract_metadata. Study: PMID29568058.

52. Frakes AE, Ferraiuolo L, Haidet-Phillips AM et al. Microglia induce motor neuron death via the classical NF-κB pathway in amyotrophic lateral sclerosis. Neuron. 2014. [PMID 24607225](https://pubmed.ncbi.nlm.nih.gov/24607225/). DOI: 10.1016/j.neuron.2014.01.013. Access: full_text_partial. Study: PMID24607225.

53. Heneka MT, Kummer MP, Stutz A et al. NLRP3 is activated in Alzheimer's disease and contributes to pathology in APP/PS1 mice. Nature. 2013. [PMID 23254930](https://pubmed.ncbi.nlm.nih.gov/23254930/). DOI: 10.1038/nature11729. Access: abstract_metadata. Study: PMID23254930.

54. Spangenberg EE, Lee RJ, Najafi AR et al. Eliminating microglia in Alzheimer's mice prevents neuronal loss without modulating amyloid-β pathology. Brain : a journal of neurology. 2016. [PMID 26921617](https://pubmed.ncbi.nlm.nih.gov/26921617/). DOI: 10.1093/brain/aww016. Access: abstract_metadata. Study: PMID26921617.

55. Guerreiro R, Wojtas A, Bras J et al. TREM2 variants in Alzheimer's disease. The New England journal of medicine. 2013. [PMID 23150934](https://pubmed.ncbi.nlm.nih.gov/23150934/). DOI: 10.1056/NEJMoa1211851. Access: abstract_metadata. Study: PMID23150934.

56. Jonsson T, Stefansson H, Steinberg S et al. Variant of TREM2 associated with the risk of Alzheimer's disease. The New England journal of medicine. 2013. [PMID 23150908](https://pubmed.ncbi.nlm.nih.gov/23150908/). DOI: 10.1056/NEJMoa1211103. Access: abstract_metadata. Study: PMID23150908.

57. Sims R, van der Lee SJ, Naj AC et al. Rare coding variants in PLCG2, ABI3, and TREM2 implicate microglial-mediated innate immunity in Alzheimer's disease. Nature genetics. 2017. [PMID 28714976](https://pubmed.ncbi.nlm.nih.gov/28714976/). DOI: 10.1038/ng.3916. Access: abstract_metadata. Study: PMID28714976.

58. ADAPT Research Group, Lyketsos CG, Breitner JC et al. Naproxen and celecoxib do not prevent AD in early results from a randomized controlled trial. Neurology. 2007. [PMID 17460158](https://pubmed.ncbi.nlm.nih.gov/17460158/). DOI: 10.1212/01.wnl.0000260269.93245.d2. Access: abstract_metadata. Study: ADAPT-original-randomization.

59. Alzheimer's Disease Anti-inflammatory Prevention Trial Research Group. Results of a follow-up study to the randomized Alzheimer's Disease Anti-inflammatory Prevention Trial (ADAPT). Alzheimer's & dementia : the journal of the Alzheimer's Association. 2013. [PMID 23562431](https://pubmed.ncbi.nlm.nih.gov/23562431/). DOI: 10.1016/j.jalz.2012.11.012. Access: abstract_metadata. Study: ADAPT-original-randomization.

60. Aisen PS, Schafer KA, Grundman M et al. Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial. JAMA. 2003. [PMID 12783912](https://pubmed.ncbi.nlm.nih.gov/12783912/). DOI: 10.1001/jama.289.21.2819. Access: abstract_metadata. Study: PMID12783912.

61. Potter H, Woodcock JH, Boyd TD et al. Safety and efficacy of sargramostim (GM-CSF) in the treatment of Alzheimer's disease. Alzheimer's & dementia (New York, N. Y.). 2021. [PMID 33778150](https://pubmed.ncbi.nlm.nih.gov/33778150/). DOI: 10.1002/trc2.12158. Access: full_text_partial. Study: PMID33778150.

62. Wang T, Perera ND, Chiam MDF et al. Necroptosis is dispensable for motor neuron degeneration in a mouse model of ALS. Cell death and differentiation. 2020. [PMID 31745214](https://pubmed.ncbi.nlm.nih.gov/31745214/). DOI: 10.1038/s41418-019-0457-8. Access: abstract_metadata. Study: PMID31745214.

63. Lepore AC, Rauck B, Dejea C et al. Focal transplantation-based astrocyte replacement is neuroprotective in a model of motor neuron disease. Nature neuroscience. 2008. [PMID 18931666](https://pubmed.ncbi.nlm.nih.gov/18931666/). DOI: 10.1038/nn.2210. Access: abstract_metadata. Study: PMID18931666.

64. Audet JN, Gowing G, Paradis R et al. Ablation of proliferating cells in the CNS exacerbates motor neuron disease caused by mutant superoxide dismutase. PloS one. 2012. [PMID 22523565](https://pubmed.ncbi.nlm.nih.gov/22523565/). DOI: 10.1371/journal.pone.0034932. Access: abstract_metadata. Study: PMID22523565.

65. Dermentzaki G, Politi KA, Lu L et al. Deletion of Ripk3 Prevents Motor Neuron Death In Vitro but not In Vivo. eNeuro. 2019. [PMID 30815534](https://pubmed.ncbi.nlm.nih.gov/30815534/). DOI: 10.1523/ENEURO.0308-18.2018. Access: abstract_metadata. Study: PMID30815534.

66. Kang J, Rivest S et al. MyD88-deficient bone marrow cells accelerate onset and reduce survival in a mouse model of amyotrophic lateral sclerosis. The Journal of cell biology. 2007. [PMID 18086918](https://pubmed.ncbi.nlm.nih.gov/18086918/). DOI: 10.1083/jcb.200705046. Access: abstract_metadata. Study: PMID18086918.

67. Guttenplan KA, Weigel MK, Adler DI et al. Knockout of reactive astrocyte activating factors slows disease progression in an ALS mouse model. Nature communications. 2020. [PMID 32719333](https://pubmed.ncbi.nlm.nih.gov/32719333/). DOI: 10.1038/s41467-020-17514-9. Access: full_text_partial. Study: PMID32719333.

68. Chen X, Firulyova M, Manis M et al. Microglia-mediated T cell infiltration drives neurodegeneration in tauopathy. Nature. 2023. [PMID 36890231](https://pubmed.ncbi.nlm.nih.gov/36890231/). DOI: 10.1038/s41586-023-05788-0. Access: full_text_partial. Study: PMID36890231.

69. Leyns CEG, Ulrich JD, Finn MB et al. TREM2 deficiency attenuates neuroinflammation and protects against neurodegeneration in a mouse model of tauopathy. Proceedings of the National Academy of Sciences of the United States of America. 2017. [PMID 29073081](https://pubmed.ncbi.nlm.nih.gov/29073081/). DOI: 10.1073/pnas.1710311114. Access: abstract_metadata. Study: PMID29073081.

## Source-specific evidence cards

### [1] Single-cell sequencing of human midbrain reveals glial activation and a Parkinson-specific neuronal state.

**Shows:** Single-nucleus profiles of six PD and five control midbrains identified activated microglial and astrocytic states alongside a dysfunctional neuronal cluster.

**Does not show:** Postmortem states and risk enrichment do not identify initiating direction or glial necessity; the central-mechanism language exceeds selective intervention.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

### [2] Astrocytic interleukin-3 programs microglia and limits Alzheimer's disease.

**Shows:** Astrocytic IL-3 programs microglial motility and aggregate clearance and restricts pathology and cognitive decline in tested AD models.

**Does not show:** Protective activation contradicts treating activation as uniformly toxic; constitutive cytokine and model backgrounds limit natural human mediation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [3] Integrated multimodal cell atlas of Alzheimer's disease.

**Shows:** An 84-donor multimodal atlas reconstructed an early phase with inflammatory glia and selective inhibitory-neuron loss before broader later neuronal loss.

**Does not show:** Pseudoprogression is cross-sectional reconstructed time, not repeated donor observation; early glia already coexist with pathology and cell loss. UpdateOf is a preceding preprint.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID37292694. Subclaims: SC1, SC4.

### [4] Parkin regulates microglial NLRP3 and represses neurodegeneration in Parkinson's disease.

**Shows:** Parkin loss increased LPS-triggered microglial NLRP3 assembly, motor impairment and dopaminergic loss; Parkin overexpression and HSP90 inhibition interrogated NLRP3 degradation.

**Does not show:** Global Parkin deletion, imposed systemic LPS and broad HSP90 effects do not isolate a microglial-only natural PD initiator.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [5] Microglia-specific IL-10 gene delivery inhibits neuroinflammation and neurodegeneration in a mouse model of Parkinson's disease.

**Shows:** Microglia-specific IL-10 delivery protected dopaminergic neurons in SNCA-overexpressing mice while inducing an activated prophagocytic microglial state and lymphocyte regulation.

**Does not show:** Activation remains present in a protective state; local cytokine effects extend beyond microglia, and imposed SNCA expression restricts scope.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [6] Block of A1 astrocyte conversion by microglia is neuroprotective in models of Parkinson's disease.

**Shows:** NLY01 reduced dopaminergic loss and behavioral deficits after synuclein fibrils and prolonged survival in A53T mice, associated with reduced neurotoxic astrocyte conversion.

**Does not show:** GLP1R agonism and forced/transgenic proteinopathy do not establish selective glial mediation or human efficacy; the clinical trial is assessed separately.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [7] Inflammasome inhibition prevents α-synuclein pathology and dopaminergic neurodegeneration in mice.

**Shows:** MCC950 inhibited NLRP3 activity and improved motor, dopamine-neuron and aggregate endpoints in several rodent PD models; degeneration itself could activate the pathway.

**Does not show:** This supports an injury-triggered causal amplifier, not obligatory inflammation-first origin; drug/pathway and induced-model scope must be retained.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [8] Complement C1q-dependent excitatory and inhibitory synapse elimination by astrocytes and microglia in Alzheimer's disease mouse models.

**Shows:** C1q deletion reduced microglial and astrocytic synapse engulfment and restored density in TauP301S; astrocytic engulfment compensated when Trem2-dependent microglial phagocytosis failed.

**Does not show:** Synapse density is not total neuronal survival, and global complement deletion affects multiple functions and engineered backgrounds.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [9] Systemic LPS causes chronic neuroinflammation and progressive neurodegeneration.

**Shows:** A single high-dose systemic LPS or TNF exposure caused sustained brain inflammation and delayed progressive TH-positive nigral-neuron loss, with TNF-receptor comparisons.

**Does not show:** The imposed systemic inflammatory insult and receptor-wide deletion do not establish native chronic glial primacy; TH loss is not an audited all-neuron count.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

### [10] Astrocytes Protect Human Dopaminergic Neurons from α-Synuclein Accumulation and Propagation.

**Shows:** Human iPSC astrocytes reduced neuronal synuclein accumulation and transfer through lysosomal clearance; ATP13A2-deficient astrocytes lost part of this protection.

**Does not show:** Protective glial function is demonstrated in culture, not intact organismal progression or a general absence of harmful glial states.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [11] Cx3cr1-deficiency exacerbates alpha-synuclein-A53T induced neuroinflammation and neurodegeneration in a mouse model of Parkinson's disease.

**Shows:** Cx3cr1 deficiency worsened AAV-A53T synuclein inflammation and dopaminergic degeneration, while wild-type synuclein injury was similar across receptor genotypes.

**Does not show:** Receptor loss can remove restraint rather than simply reduce activation; developmental and mutant-expression contexts limit broad extrapolation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [12] Complement and microglia mediate early synapse loss in Alzheimer mouse models.

**Shows:** C1q, C3 and CR3 inhibition reduced early amyloid-associated synapse loss; C1q was required for oligomer-induced synaptic and LTP toxicity in the tested models.

**Does not show:** Early means before overt plaques, not before all amyloid or neuronal stress; complementary pathway controls support conditional mediation without proving human initiation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

### [13] TREM2 Maintains Microglial Metabolic Fitness in Alzheimer's Disease.

**Shows:** TREM2 deficiency impaired microglial metabolism, and cyclocreatine restored plaque clustering while reducing adjacent neuronal dystrophy.

**Does not show:** A protective energetic/clearance program can be activated; metabolic rescue is broader than a single inflammatory mediator and occurs in amyloid-model scope.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [14] TREM2 drives microglia response to amyloid-β via SYK-dependent and -independent pathways.

**Shows:** Microglial SYK deficiency impaired plaque encasement and worsened pathology and behavior; CLEC7A activation rescued microglial activation in TREM2-R47H mice.

**Does not show:** Different activated programmes have different effects; antibody rescue of activation is not independently established complete human clinical rescue. Original Results: inducible SYK deletion lowers plaque-associated microgliosis but increases Aβ, neuritic dystrophy and memory impairment. Monocytes recover after washout; border-associated macrophages retain partial deletion. Late deletion also impairs clustering. CLEC7A rescue restores uptake and plaque morphology, not demonstrated long-term behavioral rescue.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:RESULTS:paragraph0, 1:RESULTS:paragraph2, 1:RESULTS:paragraph3, 1:RESULTS:paragraph6, 1:RESULTS:paragraph14. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [15] TREM2 receptor protects against complement-mediated synaptic loss by binding to complement C1q during neurodegeneration.

**Shows:** TREM2-C1q binding restrained complement activation; haploinsufficiency worsened synapse engulfment and a binding peptide rescued synaptic impairment in tested models.

**Does not show:** Human tissue complex associations do not establish causal direction; complement-restraint protection is distinct from globally harmful activation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [16] The neuronal pentraxin Nptx2 regulates complement activity and restrains microglia-mediated synapse loss in neurodegeneration.

**Shows:** Nptx2 loss increased C1q-dependent microglial synapse removal, while neuronal Nptx2 overexpression restrained complement and synapse loss in aged tau mice.

**Does not show:** The initiating alteration is neuronal complement restraint and one rescue uses AAV overexpression; synapse preservation is not whole-disease neuronal survival.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [17] TREM2 lipid sensing sustains the microglial response in an Alzheimer's disease model.

**Shows:** TREM2 loss impaired damage-lipid sensing, microglial plaque clustering and survival and increased amyloid accumulation in 5xFAD.

**Does not show:** Protective immune responsiveness is demonstrated in an amyloid context; amyloid load alone does not establish cognitive/neuron-loss mediation or chronic-state universality.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [18] LC3-Associated Endocytosis Facilitates β-Amyloid Clearance and Mitigates Neurodegeneration in Murine Alzheimer's Disease.

**Shows:** Myeloid or microglial LANDO loss increased cytokines and neurotoxic amyloid, with accelerated degeneration, altered signaling and memory deficits.

**Does not show:** Loss of aggregate clearance may precede harmful inflammation, so pure inflammatory primacy is not isolated; a linked correction remains qualified.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID33306957. Subclaims: SC2, SC3.

### [19] Impact of TREM2R47H variant on tau pathology-induced gliosis and neurodegeneration.

**Shows:** Human TREM2-R47H in PS19 reduced microglial reactivity, synapse phagocytosis, atrophy and synapse loss versus the common variant.

**Does not show:** An AD-risk allele is protective in this tau-only model, mapping upstream substrate/stage context rather than proving uniform effects in natural AD.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [20] Anti-Neuroinflammatory and Anti-Inflammatory Effects of the NLRP3 Inhibitor NT-0796 in Subjects with Parkinson's Disease.

**Shows:** Open-label NT-0796 in ten PD participants for 28 days showed CSF exposure and reduced inflammatory markers, with four healthy volunteers also studied.

**Does not show:** Short safety/pharmacokinetic biomarker engagement without randomized clinical trajectory is not disease modification or evidence of a glial initiating event.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [21] Complement 3-deficient mice are not protected against MPTP-induced dopaminergic neurotoxicity.

**Shows:** C3 deficiency did not preserve dopamine-neuron/fiber or dopamine-metabolite endpoints after acute or chronic MPTP exposure.

**Does not show:** This null constrains complement C3 in a toxin context, not every immune mediator; the linked notice corrects an author name.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMIDnot supplied. Subclaims: SC2, SC3.

### [22] Terminal complement pathway activation drives synaptic loss in Alzheimer's disease models.

**Shows:** C7 antibody and C6 knockout reduced plaque-adjacent spine/synapse loss in two AD models, implicating terminal complement activity.

**Does not show:** Synapse endpoints are not audited neuron-survival/function trajectories, and constitutive knockout or systemic antibody may affect other compartments.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [23] Microglia-specific overexpression of α-synuclein leads to severe dopaminergic neurodegeneration by phagocytic exhaustion and oxidative toxicity.

**Shows:** Microglia-selective synuclein overexpression caused dopamine-neuron degeneration without neuronal synuclein aggregation, and oxidative/nitrosative inhibition attenuated damage.

**Does not show:** A strong imposed glial sufficiency setting is not physiological endogenous primacy; peripheral recruitment and broad oxidative inhibition remain alternative mediators. Linked correction qualified.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID34916524. Subclaims: SC2, SC3.

### [24] STING mediates neurodegeneration and neuroinflammation in nigrostriatal α-synucleinopathy.

**Shows:** STING activation preceded dopamine-neuron loss after synuclein fibrils; STING-deficient mice resisted motor, aggregate and neuronal-loss phenotypes.

**Does not show:** Global STING deficiency and imposed fibrils constrain cell specificity and natural initiation; human staining is correlational.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

### [25] Selective removal of astrocytic APOE4 strongly protects against tau-mediated neurodegeneration and decreases synaptic phagocytosis by microglia.

**Shows:** Inducible astrocytic APOE4 removal after tau pathology onset reduced neurodegeneration, synapse loss and microglial synaptic phagocytosis.

**Does not show:** Post-onset protection supports causal amplification; APOE affects lipid and tau pathways in addition to inflammation, and the P301S/APOE context remains explicit. Original Results: astrocytic APOE deletion at 5.5 months after initial tau pathology protects APOE4 but not APOE3 mice from atrophy. Tamoxifen-treated Cre-negative controls exclude tamoxifen alone; NfL, nest building and synaptic measures improve. APOE influences lipid handling and tau pathology as well as inflammatory states, so the mediation is not inflammation-selective.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:RESULTS:paragraph0, 1:RESULTS:paragraph1, 1:RESULTS:paragraph2, 1:RESULTS:paragraph15. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [26] Phagocytic clearance of presynaptic dystrophies by reactive astrocytes in Alzheimer's disease.

**Shows:** Microscopy in APP/PS1 and human AD tissue showed reactive astrocytes engulfing presynaptic dystrophies, with a low fraction of total dystrophies cleared.

**Does not show:** Reactive phagocytosis may clear damage rather than kill viable synapses; proposed protection is not tested by this static study.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [27] Microglial Depletion with Clodronate Liposomes Increases Proinflammatory Cytokine Levels, Induces Astrocyte Activation, and Damages Blood Vessel Integrity.

**Shows:** Intracerebral clodronate depleted microglia but also damaged vascular integrity, reduced neural markers and increased inflammatory/glial signals; neuronal ATP fell in culture.

**Does not show:** This methodological perturbation exposes depletion off-target confounding, rather than showing that all microglia protect in AD, PD or ALS.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [28] Neurotoxic reactive astrocytes are induced by activated microglia.

**Shows:** Microglial IL1a, TNF and C1q induced toxic reactive astrocytes; blocking formation protected axotomized neurons, and human disease tissue contained the state.

**Does not show:** Injury precedes the causal astrocyte response in the in-vivo test; human postmortem abundance and a marker class do not establish obligatory inflammation-first disease origin.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

### [29] A Unique Microglia Type Associated with Restricting Development of Alzheimer's Disease.

**Shows:** Single-cell mapping and Trem2 loss separated two disease-associated microglial activation phases, with intracellular amyloid in mapped cells.

**Does not show:** Potential protective function is not a selective hard-outcome rescue in this abstract; the DAM name is not a causal toxicity classification.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [30] Microglia rescue neurons from aggregate-induced neuronal dysfunction and death through tunneling nanotubes.

**Shows:** Microglial nanotube contacts removed neuronal tau/synuclein aggregates, supplied healthy mitochondria and restored activity; mitochondrial poisoning removed protection.

**Does not show:** Broad mitochondrial perturbation and culture-based transfer do not establish organismal disease trajectory; aggregate transfer can be protective rather than uniformly toxic.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [31] NLRP3 inflammasome activation drives tau pathology.

**Shows:** NLRP3 loss reduced tau phosphorylation/aggregation, and amyloid-homogenate-induced tau pathology depended on NLRP3; tau itself activated the inflammasome.

**Does not show:** The test supports a bidirectional proteinopathy-immune pathway, with imposed initiation and pathology endpoints rather than a complete human survival dependency.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [32] Early microglial and astrocyte reactivity in preclinical Alzheimer's disease.

**Shows:** In 211 cognitively unimpaired people, glial fluid markers were elevated in preclinical AD and structural-equation relations linked them with amyloid, tau and injury markers.

**Does not show:** Statistical arrows and biomarker mediation do not establish causal intervention or true temporal lineage; normal cognition does not mean absence of primary protein pathology.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

### [33] Microglial PD-1 stimulation by astrocytic PD-L1 suppresses neuroinflammation and Alzheimer's disease pathology.

**Shows:** Microglial PD1 deletion increased inflammation and impaired amyloid uptake, with more plaque deposition in APP/PS1.

**Does not show:** Loss of immune restraint and clearance is harmful in this context; plaque load is not independently demonstrated neuron loss or general inflammatory primacy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [34] Microglia-derived ASC specks cross-seed amyloid-β in Alzheimer's disease.

**Shows:** ASC specks promoted amyloid cross-seeding after hippocampal delivery, and ASC deficiency or antibody coapplication blocked induced pathology.

**Does not show:** Forced amyloid/speck entry and aggregate endpoints establish inflammatory seeding capability, not native disease initiation or clinical benefit.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [35] Anti-human TREM2 induces microglia proliferation and reduces pathology in an Alzheimer's disease model.

**Shows:** AL002c stimulated microglial proliferation and reduced filamentous plaques and neurite dystrophy in a human-TREM2 AD model; first-human fluid engagement was also reported.

**Does not show:** Activation can be protective in this model, and phase1 engagement is not human clinical efficacy; later reports overlap the clinical programme.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC4.

### [36] TREM2-independent microgliosis promotes tau-mediated neurodegeneration in the presence of ApoE4.

**Shows:** TREM2 knockout worsened tau pathology and degeneration in P301S/APOE4 despite reduced TREM2-dependent microgliosis.

**Does not show:** This reverses the earlier mouse-Apoe tau-only protection contrast, showing genotype and residual microglial state matter more than bulk activation reduction.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [37] Preclinical and first-in-human evaluation of AL002, a novel TREM2 agonistic antibody for Alzheimer's disease.

**Shows:** INVOKE1 in 64 healthy volunteers showed AL002 CSF penetration, soluble-TREM2 reduction and signaling/recruitment biomarker increases with acceptable single-dose tolerability.

**Does not show:** Healthy phase1 safety and engagement cannot establish AD trajectory benefit; the 2020 programme report is not independent efficacy evidence.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [38] The Alzheimer's disease-associated protective Plcγ2-P522R variant promotes immune functions.

**Shows:** The AD-protective PLCG2-P522R knock-in increased enzyme activity, macrophage survival/acute responses and measures of microglial function.

**Does not show:** Protective genetic association is from a separate study; cell function and PET signals do not prove chronic harmful activation or disease rescue in this KI experiment.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC4.

### [39] Efficacy of minocycline in patients with amyotrophic lateral sclerosis: a phase III randomised trial.

**Shows:** In 412 randomized ALS patients, minocycline accelerated the prespecified ALSFRS-R decline compared with placebo over nine months.

**Does not show:** Broad drug pharmacology and unverified selective glial engagement mean this refutes the tested treatment expectation, not all inflammatory causal contribution; mortality worsening was nonsignificant.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [40] The TREM2 agonistic antibody AL002 in early Alzheimer's disease: a phase 2 randomized trial.

**Shows:** INVOKE2 randomized 381 early-AD participants and found no CDR-SB benefit despite sustained central TREM2 engagement/signaling biomarkers; ARIA-like MRI events were prominent.

**Does not show:** TREM2 agonism in symptomatic AD is a specific strategy; fluid engagement does not prove adequate state-specific glial rescue, and a clinical null does not eliminate all immune causality.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [41] Efficacy and safety of oral semaglutide 14 mg (flexible dose) in early-stage symptomatic Alzheimer's disease (evoke and evoke+): two phase 3, randomised, placebo-controlled trials.

**Shows:** Two phase3 semaglutide trials randomized 3808 early-AD participants and showed no slowing of 104-week CDR-SB progression.

**Does not show:** GLP1R agonism is systemic and pleiotropic, without verified selective glial engagement here; this constrains broad drug translation rather than isolating a glial mechanism.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [42] Safety, tolerability, and efficacy of NLY01 in early untreated Parkinson's disease: a randomised, double-blind, placebo-controlled trial.

**Shows:** NLY01 randomized 255 early untreated PD participants and showed no 36-week motor/ADL primary endpoint benefit at either dose, with frequent gastrointestinal events.

**Does not show:** Exploratory younger-patient signals are not confirmatory; human microglial-state inhibition was not verified by this abstract, so null treatment efficacy is not complete mediator removal.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [43] Azathioprine for the treatment of early Parkinson's disease (AZA-PD): a randomised, double-blind, placebo-controlled, proof-of-concept, phase 2 trial.

**Shows:** AZA-PD randomized 66 early PD participants and did not improve the prespecified 12-month off-state gait-axial endpoint; exploratory immune and sex-specific findings were reported.

**Does not show:** Peripheral broad immunosuppression is not cell-state-selective glial blockade, and subgroup signals require replication; clinical null constrains the tested strategy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [44] Evaluation of the safety and immunomodulatory effects of sargramostim in a randomized, double-blind phase 1 clinical Parkinson's disease trial.

**Shows:** A 20-person randomized PD phase1 study reported modest short-term motor improvement with sargramostim, peripheral Treg changes and substantial adverse-event burden.

**Does not show:** Safety-focused small enrollment, short exposure and peripheral modulation do not establish sustained neuroprotection or prove a CNS glial mediator.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [45] Heterozygous Tbk1 loss has opposing effects in early and late stages of ALS in mice.

**Shows:** Heterozygous TBK1 loss advanced SOD1 onset/denervation but reduced late microglial inflammation, slowed progression and extended survival.

**Does not show:** Global pleiotropic autophagy/immune effects show stage-dependent direction, rather than a universal beneficial inflammatory suppression or native sporadic ALS cause.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [46] System xC- is a mediator of microglial function and its deletion slows symptoms in amyotrophic lateral sclerosis mice.

**Shows:** xCT deletion reduced microglial glutamate and inflammatory outputs and slowed late SOD1 degeneration with more surviving motor neurons, despite earlier symptom onset.

**Does not show:** Global antiporter deletion and redox effects complicate microglial specificity; onset and progression move in opposite directions.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [47] C1q induction and global complement pathway activation do not contribute to ALS toxicity in mutant SOD1 mice.

**Shows:** C1q or C3 loss did not change SOD1 disease onset/progression; C1q deletion enhanced rather than prevented loss of certain motor-neuron synaptic inputs.

**Does not show:** Complement activation can be a correlate without disease-course necessity in this setting; null single mediators do not disprove all glial contributions.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [48] Astrocytes as determinants of disease progression in inherited amyotrophic lateral sclerosis.

**Shows:** Conditional mutant-SOD1 reduction in astrocytes left onset unchanged but delayed microglial activation and sharply slowed later progression.

**Does not show:** This is direct cell-resolved causal amplification in familial SOD1 models, not proof of inflammation-first initiation or every reactive astrocyte being toxic.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [49] Onset and progression in inherited ALS determined by motor neurons and microglia.

**Shows:** Cell-resolved mutant-SOD1 deletion separated motor-neuron-determined onset/early progression from microglial effects on late progression.

**Does not show:** Different disease phases and engineered familial mutation define scope; reducing mutant protein may restore several glial functions beyond inflammation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [50] Astrocyte-derived TGF-β1 accelerates disease progression in ALS mice by interfering with the neuroprotective functions of microglia and T cells.

**Shows:** Astrocytic TGFβ1 overproduction worsened SOD1 progression while deactivating protective microglia/T cells; post-onset signaling inhibition extended survival.

**Does not show:** Reducing one cytokine can improve protective immune activity rather than uniformly suppress activation; source overproduction and drug pleiotropy remain limits.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [51] Innate immune adaptor TRIF deficiency accelerates disease progression of ALS mice with accumulation of aberrantly activated astrocytes.

**Shows:** TRIF deficiency shortened SOD1 survival with accumulated abnormal astrocytes; manipulating selected lymphocyte populations did not substantially alter survival.

**Does not show:** Innate signaling can protect the motor-neuron environment, and cell correlations do not by themselves prove which abnormal astrocyte state mediates the whole effect.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [52] Microglia induce motor neuron death via the classical NF-κB pathway in amyotrophic lateral sclerosis.

**Shows:** Microglial NFκB suppression extended ALS-mouse survival, whereas astrocytic suppression did not rescue motor-neuron death; microglial activation caused death in culture and in vivo.

**Does not show:** The endogenous pathway perturbation is cell-resolved, but forced constitutive activation and SOD1 genotype restrict broad human or inflammation-first inference. Original Results: CSF1R-Cre acts in the mononuclear phagocyte lineage, although spinal reporter staining is restricted to microglia. Homozygous IKKβ deletion caused serious immune dysfunction, so survival was assessed in heterozygotes. Median survival rose from 133 to 153 days, onset was unchanged, and mutant SOD1 abundance was unchanged. Forced NFκB activation also kills neurons, but is not physiological initiation.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:RESULTS:paragraph8, 1:RESULTS:paragraph9, 1:RESULTS:paragraph12, 1:RESULTS:paragraph13. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [53] NLRP3 is activated in Alzheimer's disease and contributes to pathology in APP/PS1 mice.

**Shows:** NLRP3 or caspase1 loss preserved spatial memory and increased amyloid clearance in APP/PS1, with reduced inflammatory activity and amyloid burden.

**Does not show:** Constitutive global immune loss and familial amyloid background are limited scopes; downstream response to amyloid can still mediate causal decline.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [54] Eliminating microglia in Alzheimer's mice prevents neuronal loss without modulating amyloid-β pathology.

**Shows:** One-month CSF1R inhibition in ten-month 5xFAD removed most microglia and preserved neuronal/spine endpoints and contextual memory without reducing amyloid.

**Does not show:** This is a post-pathology causal contribution; systemic receptor pharmacology and total-cell removal are broader than a single harmful state, and protective contexts elsewhere remain.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [55] TREM2 variants in Alzheimer's disease.

**Shows:** Original discovery and replication genotyping plus GWAS meta-analysis linked rare TREM2 variants to AD risk.

**Does not show:** This is a primary genetic cohort report despite its meta-analysis component; genetic immune dysfunction does not identify uniformly harmful chronic activation or a specific inflammatory mediator.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [56] Variant of TREM2 associated with the risk of Alzheimer's disease.

**Shows:** Icelandic sequencing/imputation and international replication associated TREM2-R47H with about threefold AD risk and poorer cognition in unaffected elderly carriers.

**Does not show:** The variant's immune/clearance mechanism needs separate functional identification; protective microglial failure is as relevant as excess inflammation, and replication cohorts may overlap other reports.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [57] Rare coding variants in PLCG2, ABI3, and TREM2 implicate microglial-mediated innate immunity in Alzheimer's disease.

**Shows:** Three-stage genotyping in 85133 participants identified protective PLCG2-P522R and risk-increasing ABI3/TREM2 coding variants enriched in microglial pathways.

**Does not show:** The genetic evidence supports immune-function involvement, while protection from enhanced PLCG2 function contradicts equating activation magnitude with harm; mechanistic mediation remains separate.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [58] Naproxen and celecoxib do not prevent AD in early results from a randomized controlled trial.

**Shows:** Early ADAPT results found no prevention benefit from naproxen or celecoxib, with stopped treatment and unexpected prevalent cognitive disease at entry.

**Does not show:** Early imprecision and enrollment/treatment interruption limit interpretation; systemic NSAIDs do not selectively remove a defined glial state.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [59] Results of a follow-up study to the randomized Alzheimer's Disease Anti-inflammatory Prevention Trial (ADAPT).

**Shows:** ADAPT follow-up across approximately seven years found no overall AD incidence reduction after a median under 1.5 years of assigned NSAID treatment.

**Does not show:** Same randomized cohort as early ADAPT, not independent replication; delayed follow-up does not equal continuous long-term exposure or specific glial engagement.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [60] Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial.

**Shows:** In 351 mild-to-moderate AD participants, one-year naproxen or rofecoxib did not slow the ADAS-Cog primary outcome or show consistent secondary benefit.

**Does not show:** COX inhibition and systemic adverse effects do not prove selective CNS mediator blockade; results constrain these doses and symptomatic-stage strategies.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [61] Safety and efficacy of sargramostim (GM-CSF) in the treatment of Alzheimer's disease.

**Shows:** A 40-person randomized AD study found short-term MMSE improvement after three weeks of sargramostim, with peripheral immune and plasma biomarker changes.

**Does not show:** Safety-primary small study and brief cognitive effects do not establish sustained disease modification or a CNS glial mechanism; broader outcome and baseline details require original sections. Original Results and Methods: 40-person, three-week safety-primary trial. Baseline MMSE imbalance complicates the positive MMSE signal; baseline-adjusted end-of-treatment significance is lost, although the first follow-up remains positive. ADAS-Cog13 worsens at that follow-up, ADL/CDR-SB are null and the small PET subset is null. This is a mixed exploratory efficacy pattern, not established sustained benefit.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 2:RESULTS:paragraph11, 2:RESULTS:paragraph14, 2:RESULTS:paragraph16, 2:RESULTS:paragraph19, 2:RESULTS:paragraph23, 1:METHODS. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [62] Necroptosis is dispensable for motor neuron degeneration in a mouse model of ALS.

**Shows:** MLKL ablation did not change SOD1 onset, progression, survival, neuron degeneration or glial activation despite prior necroptosis-pathway claims.

**Does not show:** This is a terminal-effector null in one familial model, not exclusion of every inflammatory pathway; expression-marker accumulation is insufficient mechanism evidence.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [63] Focal transplantation-based astrocyte replacement is neuroprotective in a model of motor neuron disease.

**Shows:** Normal astrocyte-precursor transplantation in SOD1 rats reduced microgliosis, preserved motor neurons and respiratory/motor function and extended survival, partly through GLT1.

**Does not show:** Transplantation restores trophic/glutamate functions as well as inflammation; focal familial-model benefit does not isolate one harmful reactive state.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [64] Ablation of proliferating cells in the CNS exacerbates motor neuron disease caused by mutant superoxide dismutase.

**Shows:** AraC removal of proliferating CNS cells accelerated SOD1 progression while reducing microglia, progenitors and T cells.

**Does not show:** Multiple cell classes and anti-mitotic effects prevent attributing harm specifically to microglial loss, but contradict indiscriminate suppression as a universal strategy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [65] Deletion of Ripk3 Prevents Motor Neuron Death In Vitro but not In Vivo.

**Shows:** RIPK3 silencing protected cultured neurons from SOD1 astrocytes, but genetic deletion did not improve mouse behavior or neuropathology.

**Does not show:** Culture rescue failed organismal follow-through for this mediator; it cannot be generalized as proof that all glial toxicity pathways are merely reactive.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [66] MyD88-deficient bone marrow cells accelerate onset and reduce survival in a mouse model of amyotrophic lateral sclerosis.

**Shows:** MyD88-deficient marrow transplantation advanced onset and shortened survival in SOD1-G37R mice relative to competent donor cells.

**Does not show:** Marrow conditioning and peripheral/engrafted myeloid contributions differ from native resident microglial activation; immune signaling was protective in this tested context.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [67] Knockout of reactive astrocyte activating factors slows disease progression in an ALS mouse model.

**Shows:** Combined IL1a, TNF and C1q knockout extended SOD1-G93A survival and interrogated the proposed microglia-to-reactive-astrocyte pathway.

**Does not show:** Global lifelong cytokine deletion is not astrocyte-specific loss; rescue, stage timing and alternative immune effects require body checks before attributing all benefit to one state. Original Results: lifelong global IL1α/TNF/C1q triple deletion delays motor-neuron loss and denervation rather than preventing eventual disease; mutant animals reach a later endpoint with similar loss. Disease onset remains similar. This is causal progression modification with a constitutive, systemic intervention.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph2, 1:Results:paragraph3, 1:Results:paragraph4, 1:Results:paragraph5, 1:Results:paragraph7. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [68] Microglia-mediated T cell infiltration drives neurodegeneration in tauopathy.

**Shows:** Microglial or T-cell depletion blocked tau-model degeneration, while IFNγ or PDCD1 inhibition ameliorated atrophy; human AD tissue contained expanded cytotoxic T cells.

**Does not show:** Adaptive and innate immune effects form a causal model hub downstream of tau, rather than proving all human activation is an initiating driver; linked correction remains attached. Original Results: peripheral T-cell depletion from 6 to 9.5 months reduces atrophy, NfL and functional deficits versus IgG controls; IFNγ blockade begins at 7.5 months and PLX3397 at 8.5 months. PLX3397 targets CSF1R, c-kit and FLT3. Tau burden also changes, and neither intervention isolates a glial-only mediator. Original results verify better Y-maze/auditory-cue performance and baseline locomotor controls; contextual freezing was only a trend.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 2:Interaction of microglia and T cells:paragraph2, 2:Interaction of microglia and T cells:paragraph3, 3:T cell depletion prevent degeneration:paragraph0, 3:T cell depletion prevent degeneration:paragraph1. No complete-paper or supplement audit claimed. ErratumIn: PMID39415023. Subclaims: SC2, SC3.

### [69] TREM2 deficiency attenuates neuroinflammation and protects against neurodegeneration in a mouse model of tauopathy.

**Shows:** Trem2 deletion reduced microgliosis and preserved brain volume in PS19 despite similar phosphorylated/insoluble tau.

**Does not show:** This separates tau burden from immune-associated atrophy in a pure-tau background; amyloid and APOE4 experiments reverse or alter the direction, restricting generalization.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.